An Aggregation-Specific Enzyme-Linked Immunosorben
http://www.100md.com
病菌学杂志 2005年第19期
Institute of Pathology
Department of Neurosciences, School of Medicine, Case Western Reserve University, Cleveland, Ohio 44107-1712
Department of Biology and Biochemistry, Bath University, Bath, United Kingdom
Department of Neurology, Psychiatry and Pathology, New York University School of Medicine, New York, New York 10016
Institute of Microbiology, Chinese Academy of Science, Beijing 100080, People's Republic of China
Centre for Veterinary Science, Department of Clinical Veterinary Medicine, University of Cambridge, Cambridge, United Kingdom
ABSTRACT
The conversion of the normal cellular prion protein, PrPC, into the protease-resistant, scrapie PrPSc aggregate is the cause of prion diseases. We developed a novel enzyme-linked immunosorbent assay (ELISA) that is specific for PrP aggregate by screening 30 anti-PrP monoclonal antibodies (MAbs) for their ability to react with recombinant mouse, ovine, bovine, or human PrP dimers. One MAb that reacts with all four recombinant PrP dimers also reacts with PrPSc aggregates in ME7-, 139A-, or 22L-infected mouse brains. The PrPSc aggregate is proteinase K resistant, has a mass of 2,000 kDa or more, and is present at a time when no protease-resistant PrP is detectable. This simple and sensitive assay provides the basis for the development of a diagnostic test for prion diseases in other species. Finally, the principle of the aggregate-specific ELISA we have developed may be applicable to other diseases caused by abnormal protein aggregation, such as Alzheimer's disease or Parkinson's disease.
INTRODUCTION
All prion diseases are believed to share the same pathogenic mechanism based on the conversion of the normal cellular prion protein, PrPC, into the pathogenic scrapie PrP isoform, PrPSc (26, 27). The PrPC-to-PrPSc conversion is based on a change in conformation from a predominantly -helical structure to a predominantly ?-sheet structure (7, 20). An important effect of the conformational change is that while the entire PrPC is protease sensitive, the C-terminal domain of PrPSc becomes protease resistant.
The mechanism of PrPC-to-PrPSc conversion is complex and not completely understood. Two distinct models have been proposed; the first model suggests that PrPSc is a monomer which catalyzes the conversion of PrPC to PrPSc via a heterodimer interaction (10, 13). The second model proposes that PrPSc is an aggregate; it converts PrPC by serving as a nucleation center for the recruitment and polymerization of PrP (4). Irrespective of the mechanism, dimerization of either PrPC or PrPSc plays a crucial role in the conversion process (34).
A small amount of recombinant human prion protein (rHu-PrP) is present in dimeric form, involving the cysteine residue in the C terminus (14). Under certain conditions, such as low pH, recombinant mouse PrP (rMo-PrP) forms aggregates with an approximate molecular mass of 340 kDa (16). Aging of rMo-PrP under physiologic conditions also results in the formation of PrP dimers and multimers (29). PrPC dimers have also been detected in a cell line (25). Bovine PrPC monomer and dimer coexist in equilibrium in vivo; in contrast to native PrPC, rBo-PrP does not dimerize, due to the lack of N-linked glycans (18).
Antibody binding studies of PrPC and the proteinase K-resistant core of PrPSc suggest that residues 90 to 120 of PrPC and PrPSc, respectively, exhibit differences in their conformations (24). Moreover, binding of antibody to residues 133 to 157, comprising helix 1 in PrPC, inhibits prion propagation in vitro and in vivo (9, 11). In vitro studies using synthetic peptides have also identified residues 119 to 136, 166 to 179, and 200 to 223 on PrPC to be important in the conversion (12).
In vivo, PrPSc exists as aggregates referred to as prion rods or scrapie amyloid fibrils. Each prion rod has approximately 103 molecules of PrP (17, 28). PrPSc infectivity in hamster brain has a sedimentation coefficient of 40S (17, 28). In another study, it was estimated that the smallest PrPSc has a molecular mass of about 600 kDa (35). However, ionizing radiation inactivation experiments found that the minimum size of a PrPSc molecule has a molecular mass of 50 kDa, which corresponds to a PrP dimmer (1). Interestingly, infectivity of PrPSc could be separated from the amyloid properties of scrapie amyloid fibrils (36).
In this report, we describe the development of a novel enzyme-linked immunosorbent assay (ELISA) that reacts specifically with PrP dimers or PrP aggregates. We used this assay to compare dimeric PrP from four mammalian species, murine, ovine, bovine, and human. Furthermore, we describe the use of this assay to determine whether similar dimeric or distinct PrP aggregates are present in brain homogenates from normal or PrPSc-infected mice and discuss the nature of these PrP species.
MATERIALS AND METHODS
Recombinant PrP proteins. The generation of recombination PrP proteins from different mammalian species has been described elsewhere (3, 33, 39).
Anti-PrPC MAbs. The generation and characterization of anti-PrPC monoclonal antibodies (MAbs) have been described in detail (15, 40). All MAbs were affinity purified with protein G chromatography. MAbs were biotinylated using the EZ-Link sulfo-NHS-biotin kit (Pierce Endogen, Rockford, IL).
Mice. ME7, 139A, or 22L mouse-adapted scrapie strains were propagated by intracerebral injection into 7-week-old CD-1 (Prnpa) mice as previously described (31). Unless stated, all the animals were sacrificed at the terminal stage of the disease. For ME7 and 139A, this was approximately 170 days postinoculation, and for 22L it was approximately 140 days postinoculation. Sham-infected, age- and sex-matched CD-1 mice were used as controls. All animal experiments were carried out according to institutional regulations and standards.
Preparation of brain homogenate. To prepare 20% (wt/vol) total brain homogenate, individual entire brain samples were homogenized in ice-cold lysis buffer (phosphate-buffered saline [PBS] with 1% Nonidet P-40, 0.5% sodium deoxycholate, 5 mM EDTA; pH 8.0) in the presence of 1 mM phenylmethylsulfonyl fluoride (PMSF; Sigma, Missouri). If the homogenate was to be treated with proteases, PMSF was omitted from the lysis buffer. After centrifugation by microcentrifuge at 1,000 x g for 10 min, the supernatant was stored in aliquots at –80°C.
ELISA. Ninety-six-well plates (Costar, New York) were coated with affinity-purified capture MAbs (0.5 μg/well in 100 μl) at room temperature for 2 to 3 h. The coated plates were blocked with 3% bovine serum albumin (Sigma, Missouri) in PBS overnight at 4°C. Different recombinant PrP protein or 100 μl of diluted brain lysates (containing 60 μg of total brain proteins from a 20% total brain homogenate) was added to the wells. Plates were incubated at room temperature for 2 h and then washed with PBST (PBS with 0.05% Tween 20) three times before adding specific detecting biotinylated MAbs. After three additional washes with PBST, a horseradish peroxidase-strepavidin conjugate (Chemicon, California) was added to the plates and incubated for 1 h. The plates were washed three times with PBST, and 100 μl of 2,2'-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (Roche Diagnostic, Indiana) was dispensed into each well. After 15 min, the absorbance was read at 405 nm on a kinetic microplate reader (Molecular Device, California). The results presented are the averages of the duplicates, and all experiments were repeated at least three times.
Enzymatic treatment of brain homogenates. Each brain homogenate was treated with 50 μg/ml of proteinase K (Sigma, Missouri) at 37°C for 1 h. The protease was inactivated by the addition of PMSF to a final concentration of 3 mM.
SDS-PAGE and immunoblotting. Similar amounts of recombinant murine, human, bovine, or ovine PrP proteins were dissolved in 2x sample buffer and heated at 95°C for 5 min before separation on 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). For reducing conditions, 2-mercaptoethanol (5% final concentration; Sigma) was added to the sample buffer. The polyacrylamide gel was transferred onto nitrocellulose membrane and probed by MAb 8H4 or MAb 7A12. After incubation with horseradish peroxidase-conjugated goat anti-mouse immunoglobulin G Fc (Chemicon, California), the transferred PrP species were visualized using the chemiluminescence blotting system (Roche Diagnostics, New Jersey).
Sucrose gradient fractionation. Twenty percent total brain homogenate was prepared as described above, and Sarkosyl was added to a final concentration of 1%. After incubation for 30 min on ice, 0.5 ml of homogenate was loaded on a 10-to-60% step sucrose gradient in ultraclear centrifuge tubes (13 by 51 mm). Ultracentrifugation was carried out in SW55 rotors (Beckman, California) at 200,000 x g, 4°C, for 60 min. Fractions of 0.42 ml were collected from the top of the tube. In some experiments, blue dextran (molecular mass, 2,000 kDa; Sigma, Missouri) was included as a marker. To detect PrP species present in different sucrose gradient fractions by immunoblotting, 10 μl of each fraction was mixed with an equal volume of 2x SDS-PAGE buffer and heated at 95°C for 5 min before being loaded onto a 12% gel, and immunoblot assays were carried out as described earlier. For aggregation-specific ELISA, 5 μl of each fraction in 95 μl of PBS was added to the ELISA plate using protocols described earlier.
RESULTS
Rationale for the development of an aggregation-specific ELISA. In a conventional sandwich or capture ELISA, two MAbs with distinct binding epitopes are required: one MAb is immobilized on a solid phase to capture the antigen, and a second MAb that reacts with a distinct epitope is then used to detect the bound antigen. We reasoned that when PrP protein dimerizes or aggregates, some MAb binding epitopes would be buried, while other MAb binding epitopes might be present more than once (see Fig. 5, below). Accordingly, the number of epitopes available for binding will depend on the composition of the aggregate. If the PrP aggregate is a tetramer, then some MAb binding epitopes may be represented four times. We further postulated that if the epitope were present more than once, we might be able to use the same MAb for capture as well as detecting MAb. In addition, since dimers and aggregates are physically larger with repeating epitopes, they will have a stronger avidity with higher probability of being captured by the immobilized antibody, compared to a smaller PrP monomer. Therefore, in our dimer or aggregation-specific ELISA, one MAb is used as the capture MAb as well as the detecting MAb.
Whether all recombinant PrP (rPrP) proteins contain dimer is unresolved (14, 18, 29). We first confirmed that PrP dimer is present in rHu-PrP, rMo-PrP, rOv-PrP, and rBo-PrP proteins by immunoblotting. Using anti-PrP MAb 7A12 for detection, rHu-PrP, rMo-PrP, and rOv-PrP migrate as a 24- to 25-kDa protein with a small amount of dimeric PrP migrating as a 50-kDa protein (Fig. 1A, left panel). On the other hand, rBo-PrP migrates slower due to the presence of an additional octapeptide repeat. All dimeric rPrP contains a disulfide bond, because under reducing conditions only the 24- to 25-kDa monomeric rPrP is detected (Fig. 1A, right panel). Based on densitometry of the bands, the amount of dimeric rPrP is usually less than 5%. MAb 7A12 did not detect any rPrP with a molecular mass larger than a dimer.
Subsequently, we determined whether any of our 30 anti-PrP MAbs could react specifically with rMo-PrP dimer. The locales of some of the MAb binding epitopes are diagrammatically presented in Fig. 1B. Some MAbs react with linear sequences, while others react with conformational epitopes. Only five MAbs, 11G5 (amino acids [aa] 114 to 130), 7A12 (aa 143 to 155), 12H7 (conformational epitope from aa 145 to 200), 7C11, and 8C6 (conformational epitopes from aa 145 to 200) consistently reacted strongly with rMo-PrP (Fig. 1C). Interestingly, neither the N-terminal-specific MAb 8B4 (aa 37 to 44) nor the C-terminal-specific MAb 8F9 (aa 220 to 231) detected any dimeric rMo-PrP; therefore, both the N terminus and the C terminus are not available for binding in the rMo-PrP dimers.
Four of these MAbs, 11G5, 7A12, 12H7, and 7C11, also reacted with rBo-PrP (Fig. 1C), rOv-PrP (Fig. 1C), and rHu-PrP (Fig. 1C) to various degrees. These results suggest that these recombinant PrP dimers share certain common features that are most noticeable in the central region, which contains the MAb 11G5 (aa 114 to 130) and 7A12 (aa 143 to 155) binding epitopes. However, the availability of some of these antibody binding epitopes is species specific. For example, MAb 8C6 reacted preferentially with rMo-PrP, MAb 12A4 reacted only with rBo-PrP, MAb 8F9 reacted most strongly with rOv-PrP, and MAb 5C3 reacted more strongly with rHu-PrP. Overall, the immunoreactivity with rOv-PrP was consistently higher.
The aggregation-specific ELISA is specific for dimeric PrP. Aggregation of rMo-PrP is age dependent. Freshly prepared rMo-PrP does not contain PrP dimer (29). We prepared rMo-PrP proteins that are free of detectable dimeric PrP and rMo-PrP that contains dimer as described previously (3, 29). The presence of dimeric rMo-PrP in the preparation was first confirmed by immunoblotting with MAb 7A12. It was clear that dimeric rMo-PrP was present only in sample A (Fig. 2A) and not in sample B (Fig. 2B). We also used a conventional capture ELISA using MAb 8B4 as the immobilized, capture MAb and biotinylated MAb 7A12 as the detecting MAb to verify that the two preparations of rMo-PrP proteins had comparable amounts of protein (Fig. 2C and D).
We next determined whether the MAbs identified earlier indeed react only with dimeric rMo-PrP. When tested in the dimer-specific ELISA, MAbs 7A12 and 11G5 but not 8H4 or 8B4 reacted strongly with the rMo-PrP preparation, sample A that contains dimeric rMo-PrP, in an antigen concentration-dependent manner (Fig. 2C). On the other hand, none of the tested MAbs (11G5, 7A12, 8H4, or 8B4) was able to react with sample B, which lacks rMo-PrP dimer (Fig. 2D). Furthermore, the dimer-specific ELISA is about 100-fold more sensitive than the immunoblot assays. These results provide strong evidence that our aggregation-specific ELISA indeed reacts with rMo-PrP dimers.
Detection of PrP aggregates in PrPSc-infected mouse brains. We next determined whether the aggregation-specific ELISA could detect PrP aggregates present in brain homogenates from mice infected with the ME7 strain of PrPSc. We screened the 30 anti-PrP MAbs and compared the immunoreactivities detected in infected brains with normal sham-infected control brains. Brain homogenates from Prnp–/–-129/Ola mice were also used as negative controls. To assist the comparison, the results from rMo-PrP are also presented (Fig. 3A). None of the anti-PrP MAbs had significant immunoreactivity with brain homogenates from either sham-infected control mice or Prnp–/–-129/Ola mice (not shown). Only 4 of the 30 anti-PrP MAbs reacted strongly with brain homogenates from infected mice (Fig. 3B).
Interestingly, MAbs 8C6, 7A12, and 12H7, which reacted strongly with rMo-PrP protein, did not react with infected brain homogenates at all. On the other hand, MAbs 7H6, 6H3, and 8F9, which did not react with rMo-PrP, reacted robustly with infected brain homogenates. Only MAb 11G5 reacted with both rMo-PrP proteins and infected brain homogenates. Furthermore, the immunoreactivity differences between infected and normal brain homogenates were profound. For example, when tested with MAb 11G5, the difference in immunoreactivity (as defined by the optical density [OD]) between infected and normal brain homogenate was more than 300%. The immunoreactivity detected with MAb 11G5 was also much stronger in infected brains than in the rMo-PrP preparation. Overall, these results suggest that the nature of the PrP aggregates formed in infected brain and rMo-PrP dimer are different.
By using MAb 11G5, we found that the aggregation-specific ELISA could detect signals in an infected brain homogenate, which contains between approximately 0.6 and 6 μg of total brain proteins (Fig. 3C, left panel). Furthermore, the binding of biotinylated 11G5 was blocked by unconjugated MAb 11G5 in a MAb-specific and concentration-dependent manner. The irrelevant MAb 8B4 did not block (Fig. 3C, right panel). When ELISA plates were first coated with an irrelevant, non-anti-PrP MAb, there was no binding of biotinylated MAb 11G5 (not shown). Therefore, the aggregation-specific ELISA is both antigen and antibody specific.
The aggregation-specific ELISA is applicable to two other strains of PrPSc. We next determined whether our findings in ME7-infected brains are applicable to two other strains of PrPSc, 22L and 139A. Individual brain homogenate was prepared from ME7-infected (n = 4), 22L-infected (n = 4), and 139A-infected (n = 4) mice at the terminal stages of disease, and each homogenate was analyzed individually. We found that MAb 11G5/11G5 also reacted with PrP aggregates in mice infected with either 22L or 139A PrPSc (Table 1). Similar to ME7-infected brains, the immunoreactivity was robust and highly reproducible. Thus, the aggregate-specific assay is applicable to at least three mouse PrPSc strains.
The immunoreactivity detected by the aggregation-specific ELISA is associated with PK-resistant PrP species. One cardinal feature of PrPSc is its proteinase K (PK) resistance (2). We determined whether the binding activity detected with MAb 11G5/11G5 is associated with PK-resistant PrP species in mice infected with ME7 (n = 4), 139A (n = 4), or 22L (n = 4) PrPSc. Each brain homogenate was divided into two tubes. One was treated with PBS, and the other was treated with 50 μg/ml of PK as described earlier. PK digestion did not reduce the binding in 22L-, 139A-, or ME7-infected brains (Table 1). Therefore, the PrP aggregates detected in this assay are PK resistant and most likely represent PrPSc.
The aggregation-specific ELISA reacts with PrPSc aggregates of various sizes in infected brains. Recently, we found that after ultracentrifugation in a 10-to-60% sucrose gradient, all the mouse PrPC species are present in the top fractions, mainly in fractions 1 and 2. In sharp contrast, in PrPSc-infected mouse brain, immunoreactivity is present in all fractions, with the strongest reactivity present in the bottom fractions, fractions 10 and 11 (21). These bottom fractions are known to contain the largest PrPSc aggregates (35).
We fractionated one control brain homogenate and one ME7-infected brain homogenate in a sucrose gradient. Each fraction was then collected and divided into two samples; one sample was immunoblotted with MAb 8H4, which detects all PrP species, to document the distribution of the PrP species, and the second sample was used to detect PrP aggregate using MAb 11G5 in the aggregation-specific ELISA. Consistent with our earlier findings, in control mice, all PrPC binding activity was present in fractions 1 and 2 (Fig. 4A). In contrast, the PrP immunoreactivity in infected brain was dispersed among all fractions, with fractions 10 and 11 having the highest intensities (Fig. 4B).
We did not detect significant MAb 11G5 immunoreactivity in any of the fractions from normal control mice (Fig. 4C). In contrast, MAb 11G5 detected high levels of immunoreactivity in fractions 3, 4, and 5. On the other hand, fractions 10 and 11, which have the largest PrP aggregates, also contained MAb 11G5 immunoreactivity, but the levels were much lower (Fig. 4C). Blue dextran has a molecular mass of about 2,000 kDa, and it partitions in fraction 3 when centrifuged under identical conditions. Therefore, a majority of the PrPSc aggregates detected by this assay are PrP aggregates of heterogeneous size with molecular mass ranging from around 2,000 kDa to larger than 2,000 kDa, but these aggregates are smaller than the largest PrPSc aggregates present in fractions 10 and 11. This experiment has been repeated with three additional ME7-infected brain homogenates as well as brain homogenates from 139A-infected or 22L-infected mice with comparable results (not shown).
Detection of PrP aggregates during disease progression. ME7-inoculated CD-1 mice begin to show signs at about 130 to 160 days postinfection and die within 3 weeks. Previously, we found that PK-resistant PrP species are only detected in animals infected 140 days earlier (21). To determine when the PrP aggregates first become detectable, brain tissues were obtained from sham-infected mice (n = 4), mice infected 30 days (n = 4) or 70 days earlier (n = 4) and not exhibiting clinical signs, mice infected 140 days earlier (n = 4) and exhibiting obvious clinical signs, and mice at a terminal stage (about 170 days) of disease (n = 4). Significant PrP aggregate immunoreactivity was detected in animals 70 days postinfection, at a time when PK-resistant PrP species are undetectable by immunoblotting (21, 22) (Table 2). These results suggest PrP aggregates are detectable in infected animals much earlier than the manifestation of clinical signs and the detection of PK-resistant PrP species.
DISCUSSION
A panel of 30 different MAbs were developed against recombinant PrP (21, 40), and by screening these anti-PrP MAbs, we have identified five MAbs that preferentially react with rMo-PrP dimers in a dimer-specific ELISA (Fig. 5). Most noteworthy are MAbs 11G5 (aa 114 to 130) and 7A12 (aa 143 to 155), which also react strongly with rBo-PrP, rOv-PrP, and rHu-PrP. Therefore, the epitopes recognized by these two MAbs are conserved across these four species. On the other hand, the binding of the other three MAbs is more variable between rPrP from different species. We also identified MAbs that are species specific, which may reflect the conformational differences among the recombinant PrP or PrP dimers from these four animal species.
None of the 30 MAbs has significant binding with brain homogenates from control, sham-infected mice or Prnp–/– mice. Therefore, if dimeric PrPC is present in normal brain, it is not detected with these MAbs. Of the five MAbs that reacted with rMo-PrP dimers, only MAb 11G5 reacted strongly with brain homogenates from ME7-infected mice. On the other hand, MAb 7A12, which reacts strongly with all four tested recombinant PrP dimers, did not react with infected mouse brain homogenates. This result suggests that as a consequence of PrPC-to-PrPSc conversion the helix 1 region (aa 143 to 157) of the molecule has changed. The helix 1 region may be important in the pathogenesis of prion disease. In a cell model, helix 1 of PrP is a major determinant of PrP folding. Disruption of helix 1 prevents the attachment of the glycophosphatidylinositol anchor and the formation of the N-linked glycans (37). In the absence of the glycophosphatidylinositol anchor, helix 1 induces the formation of unglycosylated and partially protease-resistant PrP aggregates. In all sequenced PrPC with the exception of rodent PrPC, this region also contains the sequence DYEDRYYREN, which is composed entirely of hydrophilic amino acids. In rodents PrPC, the second amino acid, tyrosine (Y), is replaced with a tryptophan. It has been suggested that this region is important in the formation of the hydrophilic core and seeding of PrP aggregates (19). This region also contains the YYR epitope, which has been reported to be exposed only in PrPSc and is not available for binding in PrPC (23). Biophysical studies have also provided strong evidence that PrPC-to-PrPSc conversion involves the conversion of -helix 1 to a ?-sheet structure (6, 30, 32). Another explanation for the inability of 7A12 to detect native PrPSc aggregate may be due to the presence of N-linked glycans. The presence of N-linked glycan in PrPSc aggregates may interfere with the binding of MAb 7A12.
On the other hand, MAb 11G5 reacts with both rPrP dimers and PrPSc aggregates in infected brains. The epitope of MAb 11G5 (aa 114 to 130) includes the first ?-strand (aa 128 to 131). Hence, the conformation of this region may be similar between rPrP dimer and PrPSc aggregates, and PrPC-to-PrPSc conversion may not change the overall conformation of this region. In vitro studies using synthetic peptide have also identified residues 119 to 136 on PrPC to be important in the conversion process (12). Analysis of 27 mammalian and 9 avian PrP proteins revealed that the most conserved region outside the globular domain is located between residues 113 and 137; thus, this part of the molecule must be important in the biology of PrPC (38).
Worthy of note is that the binding of MAb 11G5 to infected brain homogenate is about 300% higher than to rMo-PrP. Since rMo-PrP dimer has two MAb 11G5-reactive epitopes, the PrP aggregates present in infected brain may contain multiple PrP molecules with more than two MAb 11G5-reactive epitopes. Furthermore, MAb 11G5 also reacts with brain homogenates from animals infected with either one of the two other strains of mouse PrPSc, namely, 139A and 22L. Hence, while different strains of PrPSc are known to have different conformations, the MAb 11G5 epitope on PrPSc aggregate is shared between three different strains of mouse PrPSc.
We also identified three MAbs, 7H6, 6H3, and 8F9, which did not react with rMo-PrP dimer but reacted significantly with infected brain homogenates. The MAb 7H6-reactive epitope (aa 130 to 140) is contiguous to the MAb 11G5-reactive epitope and right before the helix 1 region. MAb 6H3 reacts with a conformational epitope which is located at the C-terminal region (15). The MAb 6H3-reactive epitope is quite unusual, as its availability for binding is critically dependent on the N terminus (15). Previously, we reported that binding of MAb 6H3 to recombinant rHu-PrP could be blocked by the binding of MAb 8B4, which binds to the N-terminal end of rHu-PrP. Accordingly, we speculated that there might be interactions between the N terminus and C terminus of the rHu-PrP protein. However, the relationship between these observations and the presence of multiple MAb 6H3-reactive epitopes in PrPSc aggregates in infected brains is not clear.
MAb 8F9 (aa 220 to 231) does not react with rMo-PrP dimer in our dimer-specific ELISA but reacts significantly with the PrPSc aggregates in infected brains. Recently, a MAb was generated by immunizing mice with a linear sequence encompassing residues from 214 to 226 of PrP. This MAb reacts with a conformational epitope which is available for binding in PrPSc but not in PrPC or recombinant PrP (8). Both of these results suggest that the conformation of the C terminus is amenable to change during the conversion process.
By ultracentrifugation in a sucrose gradient, most of the immunoreactivity detected by the aggregation ELISA is present in fractions 3, 4, and 5. These fractions are mostly devoid of PrPC, as it partitions to the top two fractions. Interestingly, while the bottom fractions also have immunoreactivity, the levels of binding are much lower in these largest PrPSc aggregates, which indicates that this aggregate-specific ELISA is more efficient in detecting smaller aggregates than the larger one. When fractionated in the same gradient, the 2,000-kDa blue dextran is present in fraction 3. Therefore, the aggregated PrPSc being detected is likely to have a mass similar to blue dextran and larger.
At the present time, the relationship between the PrP aggregates detected by the aggregation-specific ELISA and the largest PrPSc aggregates in the infected brains is not known. Using an in vitro conversion assay, it is estimated that the "seed" or converting activity associated with PrPSc is heterogeneous in size but larger than the molecular mass standard, blue dextran (5). It should be noted that while all fractions from sham-infected mice are sensitive to PK digestion, PK-resistant PrP species are detected in all fractions from infected mice (results not shown). These results suggest that the size of PK-resistant PrP species is heterogeneous. Hence, the aggregation-specific ELISA detects only a subpopulation of the PK-resistant species. The composition of these PrPSc aggregates is not known. In addition to PrP, it is possible that they may contain other cellular components, such as nucleic acids, lipids, non-PrP proteins, polysaccharides, or glycosaminoglycans.
Our time-course studies in ME7-infected mice revealed that the appearance of the PK-resistant PrP species and clinical signs all occur around the same time, at 140 days postinoculation. On the other hand, using MAb 11G5 the accumulation of the PrPSc aggregate was detected earlier, at 70 days postinfection. We did not detect any immunoreactivity in animals infected 30 days earlier. These results are in agreement with our recent finding that by using an epitope-scanning assay to detect changes in the conformations of PrPSc, the earliest change detected was in animals infected approximately 70 days earlier (21). We also found that, at this time, aberrant full-length and truncated PrP species begin to accumulate in the infected brains (22). The PrPSc we are detecting with the aggregation ELISA is PK resistant. The reason that we can detect PrPSc at 70 days postinoculation is most likely because the ELISA is more sensitive than immunoblot assays. The assay can routinely detect significant binding in lysates containing between 0.6 and 6 μg of total brain proteins. Using rMo-PrP as standard, we estimated that the PrPC in normal brain accounts for approximately 0.01% of the total brain proteins (results not shown). Therefore, the aggregation ELISA has the potential to detect between 0.06 and 0.006 μg of aggregated PrP. However, it will require highly purified PrPSc aggregates to precisely determine the sensitivity of the aggregation-specific ELISA. So far, the accumulation of PrP aggregates during disease progression has only been carried out with MAb 11G5. MAb 11G5 is specific for an epitope (aa 115 to 130) at the central region, and this region is exposed in the recombinant PrP dimeric structure (14). The majority of the screened anti-PrP MAbs cannot react with the PrPSc aggregates at the terminal stage of disease, which may be caused by the masking of their binding sites during the progressive aggregation of PrPSc. Therefore, it is possible that by using other MAbs we may be able to detect PrP aggregates at earlier time points after infection, when PrP aggregates are smaller.
Currently, all the in vitro diagnostic tests for prion diseases require either the demonstration of PK-resistant PrP species in brain homogenate or the uncovering of hidden epitopes after GdHCl treatment (30a). The aggregation-specific ELISA described here provides certain advantages: (i) the assay is much more sensitive because of the use of a capture antibody; (ii) the assay is simpler because a denaturing step is not required; and (iii) with the use of more than one antibody, a built-in control can be used to monitor the specificity and sensitivity of the assay. While all the tests were carried out in infected mice, based on our studies with rOv-PrP, rBo-PrP, and rHu-PrP it is likely that the same approach can be used to develop a test for other animal prion diseases as well as in human prion diseases. Indeed, recent results suggest that the aggregate-specific assay also detects PrPSc in Creutzfeldt-Jacob disease patients (B. Chang et al., unpublished data). Finally, the principle of the aggregate-specific ELISA we have developed may be applicable to other diseases caused by abnormal protein aggregation, such as Alzheimer's disease or Parkinson's disease.
ACKNOWLEDGMENTS
This work is supported in part by NIH grants NS-045981-01 (to M.S.S.), AG20245 (to T.W.), and an award/contract from the U.S. Department of Army, DAMD17-03-1-0286 (to M.S.S.) and from BBSRC (8/BSD17730) to (I.M.).
We thank Witold K. Surewicz for providing the recombinant human PrP and Neil Greenspan for suggestions.
REFERENCES
Bellinger-Kawahara, C. G., E. Kempner, D. Groth, R. Gabizon, and S. B. Prusiner. 1988. Scrapie prion liposomes and rods exhibit target sizes of 55,000 Da. Virology 164:537-541.
Bolton, D. C., M. P. McKinley, and S. B. Prusiner. 1982. Identification of a protein that purifies with the scrapie prion. Science 218:1309-1311.
Brown, D. R., I. K. Iordanova, B. S. Wong, C. Venien-Bryan, F. Hafiz, L. L. Glasssmith, M. S. Sy, P. Gambetti, I. M. Jones, C. Clive, and S. J. Haswell. 2000. Functional and structural differences between the prion protein from two alleles prnpa and prnpb of mouse. Eur. J. Biochem. 267:2452-2459.
Caughey, B., and G. S. Baron. 2002. Factors affecting interactions between prion protein isoforms. Biochem. Soc. Trans. 30:565-569.
Caughey, B., D. A. Kocisko, G. J. Raymond, and P. T. Lansbury, Jr. 1995. Aggregates of scrapie-associated prion protein induce the cell-free conversion of protease-sensitive prion protein to the protease-resistant state. Chem. Biol. 2:807-817.
Cohen, F. E. 1999. Protein misfolding and prion diseases. J. Mol. Biol. 293:313-320.
Cohen, F. E., and S. B. Prusiner. 1998. Pathologic conformations of prion proteins. Annu. Rev. Biochem. 67:793-819.
Curin, S. V., M. Bresjanac, M. Popovic, H. K. Pretnar, V. Galvani, R. Rupreht, M. Cernilec, T. Vranac, I. Hafner, and R. Jerala. 2004. Monoclonal antibody against a peptide of human prion protein discriminates between Creutzfeldt-Jacob's disease-affected and normal brain tissue. J. Biol. Chem. 279:3794-3798.
Enari, M., E. Flechsig, and C. Weissmann. 2001. Scrapie prion protein accumulation by scrapie-infected neuroblastoma cells abrogated by exposure to a prion protein antibody. Proc. Natl. Acad. Sci. USA 98:9295-9299.
Griffith, J. S. 1967. Self-replication and scrapie. Nature 215:1043-1044.
Heppner, F. L., C. Musahl, I. Arrighi, M. A. Klein, T. Rulicke, B. Oesch, R. M. Zinkernagel, U. Kalinke, and A. Aguzzi. 2001. Prevention of scrapie pathogenesis by transgenic expression of anti-prion protein antibodies. Science 294:178-182.
Horiuchi, M., G. S. Baron, L. W. Xiong, and B. Caughey. 2001. Inhibition of interactions and interconversions of prion protein isoforms by peptide fragments from the C-terminal folded domain. J. Biol. Chem. 276:15489-15497.
Huang, Z., S. B. Prusiner, and F. E. Cohen. 1996. Structures of prion proteins and conformational models for prion diseases. Curr. Top. Microbiol. Immunol. 207:49-67.
Knaus, K. J., M. Morillas, W. Swietnicki, M. Malone, W. K. Surewicz, and V. C. Yee. 2001. Crystal structure of the human prion protein reveals a mechanism for oligomerization. Nat. Struct. Biol. 8:770-774.
Li, R., T. Liu, B.-S. Wong, T. Pan, M. Morillas, W. Swietnicki, K. O'Rourke, P. Gambetti, W. K. Surewicz, and M.-S. Sy. 2000. Identification of an epitope in the C terminus of normal prion protein whose expression is modulated by binding events in the N-terminus. J. Mol. Biol. 301:567-574.
Lu, B. Y., and J. Y. Chang. 2001. Isolation of isoforms of mouse prion protein with PrPSC-like structural properties. Biochemistry 40:13390-13396.
McKinley, M. P., M. B. Braunfeld, C. G. Bellinger, and S. B. Prusiner. 1986. Molecular characteristics of prion rods purified from scrapie-infected hamster brains. J. Infect. Dis. 154:110-120.
Meyer, R. K., A. Lustig, B. Oesch, R. Fatzer, A. Zurbriggen, and M. Vandevelde. 2000. A monomer-dimer equilibrium of a cellular prion protein (PrPC) not observed with recombinant PrP. J. Biol. Chem. 275:38081-38087.
Morrissey, M. P., and E. I. Shakhnovich. 1999. Evidence for the role of PrPC helix 1 in the hydrophilic seeding of prion aggregates. Proc. Natl. Acad. Sci. USA 96:11293-11298.
Pan, K. M., M. Baldwin, J. Nguyen, M. Gasset, A. Serban, D. Groth, I. Mehlhorn, Z. Huang, R. J. Fletterick, F. E. Cohen, and S. B. Prusiner. 1993. Conversion of alpha-helices into beta-sheets features in the formation of the scrapie prion proteins. Proc. Natl. Acad. Sci. USA 90:10962-10966.
Pan, T., R. Li, S. C. Kang, B. S. Wong, T. Wisniewski, and M. S. Sy. 2004. Epitope scanning reveals gain and loss of strain specific antibody binding epitopes associated with the conversion of normal cellular prion to scrapie prion. J. Neurochem. 90:1205-1217.
Pan, T., P. Wong, B. Chang, C. Li, R. Li, S. C. Kang, T. Wisniewski, and M. S. Sy. 2005. Biochemical fingerprints of prion infection: accumulations of aberrant full-length and N-terminally truncated PrP species are common features in mouse prion disease. J. Virol. 79:934-943.
Paramithiotis, E., M. Pinard, T. Lawton, S. LaBoissiere, V. L. Leathers, W. Q. Zou, L. A. Estey, J. Lamontagne, M. T. Lehto, L. H. Kondejewski, G. P. Francoeur, M. Papadopoulos, A. Haghighat, S. J. Spatz, M. Head, R. Will, J. Ironside, K. O'Rourke, Q. Tonelli, H. C. Ledebur, A. Chakrabartty, and N. R. Cashman. 2003. A prion protein epitope selective for the pathologically misfolded conformation. Nat. Med. 9:893-899.
Peretz, D., R. A. Williamson, Y. Matsunaga, H. Serban, C. Pinilla, R. B. Bastidas, R. Rozenshteyn, T. L. James, R. A. Houghten, F. E. Cohen, S. B. Prusiner, and D. R. Burton. 1997. A conformational transition at the N terminus of the prion protein features in formation of the scrapie isoform. J. Mol. Biol. 273:614-622.
Priola, S. A., B. Caughey, K. Wehrly, and B. Chesebro. 1995. A 60-kDa prion protein (PrP) with properties of both the normal and scrapie-associated forms of PrP. J. Biol. Chem. 270:3299-3305.
Prusiner, S. B. 1982. Novel proteinaceous infectious particles cause scrapie. Science 216:136-144.
Prusiner, S. B. 1998. Prions. Proc. Natl. Acad. Sci. USA 95:13363-13383.
Prusiner, S. B., M. P. McKinley, K. A. Bowman, D. C. Bolton, P. E. Bendheim, D. F. Groth, and G. G. Glenner. 1983. Scrapie prions aggregate to form amyloid-like birefringent rods. Cell 35:349-358.
Qin, K., D. S. Yang, Y. Yang, M. A. Chishti, L. J. Meng, H. A. Kretzschmar, C. M. Yip, P. E. Fraser, and D. Westaway. 2000. Copper(II)-induced conformational changes and protease resistance in recombinant and cellular PrP. Effect of protein age and deamidation. J. Biol. Chem. 275:19121.
Safar, J., P. P. Roller, D. C. Gajdusek, and C. J. Gibbs, Jr. 1993. Conformational transitions, dissociation, and unfolding of scrapie amyloid (prion) protein. J. Biol. Chem. 268:20276-20284.
Safar, J., H. Wille, V. Itri, D. Groth, H. Serban, M. Torchia, F. E. Cohen, and S. B. Prusiner. 1998. Eight prion strains have PrP(Sc) molecules with different conformations. Nat. Med. 4:1157-1165.
Sigurdsson, E. M., M. S. Sy, R. Li, H. Scholtzova, R. J. Kascsak, R. Kascsak, R. Carp, H. C. Meeker, B. Frangione, and T. Wisniewski. 2003. Anti-prion antibodies for prophylaxis following prion exposure in mice. Neurosci. Lett. 336:185-187.
Speare, J. O., T. S. Rush III, M. E. Bloom, and B. Caughey. 2003. The role of helix 1 aspartates and salt bridges in the stability and conversion of prion protein. J. Biol. Chem. 278:12522-12529.
Swietnicki, W., R. Petersen, P. Gambetti, and W. K. Surewicz. 1997. pH-dependent stability and conformation of the recombinant human prion protein PrP(90-231). J. Biol. Chem. 272:27517-27520.
Tompa, P., G. E. Tusnady, P. Friedrich, and I. Simon. 2002. The role of dimerization in prion replication. Biophys. J. 82:1711-1718.
Tzaban, S., G. Friedlander, O. Schonberger, L. Horonchik, Y. Yedidia, G. Shaked, R. Gabizon, and A. Taraboulos. 2002. Protease-sensitive scrapie prion protein in aggregates of heterogeneous sizes. Biochemistry 41:12868-12875.
Wille, H., G. F. Zhang, M. A. Baldwin, F. E. Cohen, and S. B. Prusiner. 1996. Separation of scrapie prion infectivity from PrP amyloid polymers. J. Mol. Biol. 259:608-621.
Winklhofer, K. F., J. Heske, U. Heller, A. Reintjes, W. Muranyi, I. Moarefi, and J. Tatzelt. 2003. Determinants of the in vivo folding of the prion protein. A bipartite function of helix 1 in folding and aggregation. J. Biol. Chem. 278:14961-14970.
Wopfner, F., G. Weidenhofer, R. Schneider, A. von Brunn, S. Gilch, T. F. Schwarz, T. Werner, and H. M. Schatzl. 1999. Analysis of 27 mammalian and 9 avian PrPs reveals high conservation of flexible regions of the prion protein. J. Mol. Biol. 289:1163-1178.
Yin, S. M., Y. Zheng, and P. Tien. 2003. On-column purification and refolding of recombinant bovine prion protein: using its octarepeat sequences as a natural affinity tag. Protein Expr. Purif. 32:104-109.
Zanusso, G., D. Liu, S. Ferrari, I. Hegyi, X. Yin, A. Aguzzi, S. Hornemann, S. Liemann, R. Glockshuber, J. C. Manson, P. Brown, R. B. Petersen, P. Gambetti, and M. S. Sy. 1998. Prion protein expression in different species: analysis with a panel of new mAbs. Proc. Natl. Acad. Sci. USA 95:8812-8816.(Tao Pan, Binggong Chang, )
Department of Neurosciences, School of Medicine, Case Western Reserve University, Cleveland, Ohio 44107-1712
Department of Biology and Biochemistry, Bath University, Bath, United Kingdom
Department of Neurology, Psychiatry and Pathology, New York University School of Medicine, New York, New York 10016
Institute of Microbiology, Chinese Academy of Science, Beijing 100080, People's Republic of China
Centre for Veterinary Science, Department of Clinical Veterinary Medicine, University of Cambridge, Cambridge, United Kingdom
ABSTRACT
The conversion of the normal cellular prion protein, PrPC, into the protease-resistant, scrapie PrPSc aggregate is the cause of prion diseases. We developed a novel enzyme-linked immunosorbent assay (ELISA) that is specific for PrP aggregate by screening 30 anti-PrP monoclonal antibodies (MAbs) for their ability to react with recombinant mouse, ovine, bovine, or human PrP dimers. One MAb that reacts with all four recombinant PrP dimers also reacts with PrPSc aggregates in ME7-, 139A-, or 22L-infected mouse brains. The PrPSc aggregate is proteinase K resistant, has a mass of 2,000 kDa or more, and is present at a time when no protease-resistant PrP is detectable. This simple and sensitive assay provides the basis for the development of a diagnostic test for prion diseases in other species. Finally, the principle of the aggregate-specific ELISA we have developed may be applicable to other diseases caused by abnormal protein aggregation, such as Alzheimer's disease or Parkinson's disease.
INTRODUCTION
All prion diseases are believed to share the same pathogenic mechanism based on the conversion of the normal cellular prion protein, PrPC, into the pathogenic scrapie PrP isoform, PrPSc (26, 27). The PrPC-to-PrPSc conversion is based on a change in conformation from a predominantly -helical structure to a predominantly ?-sheet structure (7, 20). An important effect of the conformational change is that while the entire PrPC is protease sensitive, the C-terminal domain of PrPSc becomes protease resistant.
The mechanism of PrPC-to-PrPSc conversion is complex and not completely understood. Two distinct models have been proposed; the first model suggests that PrPSc is a monomer which catalyzes the conversion of PrPC to PrPSc via a heterodimer interaction (10, 13). The second model proposes that PrPSc is an aggregate; it converts PrPC by serving as a nucleation center for the recruitment and polymerization of PrP (4). Irrespective of the mechanism, dimerization of either PrPC or PrPSc plays a crucial role in the conversion process (34).
A small amount of recombinant human prion protein (rHu-PrP) is present in dimeric form, involving the cysteine residue in the C terminus (14). Under certain conditions, such as low pH, recombinant mouse PrP (rMo-PrP) forms aggregates with an approximate molecular mass of 340 kDa (16). Aging of rMo-PrP under physiologic conditions also results in the formation of PrP dimers and multimers (29). PrPC dimers have also been detected in a cell line (25). Bovine PrPC monomer and dimer coexist in equilibrium in vivo; in contrast to native PrPC, rBo-PrP does not dimerize, due to the lack of N-linked glycans (18).
Antibody binding studies of PrPC and the proteinase K-resistant core of PrPSc suggest that residues 90 to 120 of PrPC and PrPSc, respectively, exhibit differences in their conformations (24). Moreover, binding of antibody to residues 133 to 157, comprising helix 1 in PrPC, inhibits prion propagation in vitro and in vivo (9, 11). In vitro studies using synthetic peptides have also identified residues 119 to 136, 166 to 179, and 200 to 223 on PrPC to be important in the conversion (12).
In vivo, PrPSc exists as aggregates referred to as prion rods or scrapie amyloid fibrils. Each prion rod has approximately 103 molecules of PrP (17, 28). PrPSc infectivity in hamster brain has a sedimentation coefficient of 40S (17, 28). In another study, it was estimated that the smallest PrPSc has a molecular mass of about 600 kDa (35). However, ionizing radiation inactivation experiments found that the minimum size of a PrPSc molecule has a molecular mass of 50 kDa, which corresponds to a PrP dimmer (1). Interestingly, infectivity of PrPSc could be separated from the amyloid properties of scrapie amyloid fibrils (36).
In this report, we describe the development of a novel enzyme-linked immunosorbent assay (ELISA) that reacts specifically with PrP dimers or PrP aggregates. We used this assay to compare dimeric PrP from four mammalian species, murine, ovine, bovine, and human. Furthermore, we describe the use of this assay to determine whether similar dimeric or distinct PrP aggregates are present in brain homogenates from normal or PrPSc-infected mice and discuss the nature of these PrP species.
MATERIALS AND METHODS
Recombinant PrP proteins. The generation of recombination PrP proteins from different mammalian species has been described elsewhere (3, 33, 39).
Anti-PrPC MAbs. The generation and characterization of anti-PrPC monoclonal antibodies (MAbs) have been described in detail (15, 40). All MAbs were affinity purified with protein G chromatography. MAbs were biotinylated using the EZ-Link sulfo-NHS-biotin kit (Pierce Endogen, Rockford, IL).
Mice. ME7, 139A, or 22L mouse-adapted scrapie strains were propagated by intracerebral injection into 7-week-old CD-1 (Prnpa) mice as previously described (31). Unless stated, all the animals were sacrificed at the terminal stage of the disease. For ME7 and 139A, this was approximately 170 days postinoculation, and for 22L it was approximately 140 days postinoculation. Sham-infected, age- and sex-matched CD-1 mice were used as controls. All animal experiments were carried out according to institutional regulations and standards.
Preparation of brain homogenate. To prepare 20% (wt/vol) total brain homogenate, individual entire brain samples were homogenized in ice-cold lysis buffer (phosphate-buffered saline [PBS] with 1% Nonidet P-40, 0.5% sodium deoxycholate, 5 mM EDTA; pH 8.0) in the presence of 1 mM phenylmethylsulfonyl fluoride (PMSF; Sigma, Missouri). If the homogenate was to be treated with proteases, PMSF was omitted from the lysis buffer. After centrifugation by microcentrifuge at 1,000 x g for 10 min, the supernatant was stored in aliquots at –80°C.
ELISA. Ninety-six-well plates (Costar, New York) were coated with affinity-purified capture MAbs (0.5 μg/well in 100 μl) at room temperature for 2 to 3 h. The coated plates were blocked with 3% bovine serum albumin (Sigma, Missouri) in PBS overnight at 4°C. Different recombinant PrP protein or 100 μl of diluted brain lysates (containing 60 μg of total brain proteins from a 20% total brain homogenate) was added to the wells. Plates were incubated at room temperature for 2 h and then washed with PBST (PBS with 0.05% Tween 20) three times before adding specific detecting biotinylated MAbs. After three additional washes with PBST, a horseradish peroxidase-strepavidin conjugate (Chemicon, California) was added to the plates and incubated for 1 h. The plates were washed three times with PBST, and 100 μl of 2,2'-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (Roche Diagnostic, Indiana) was dispensed into each well. After 15 min, the absorbance was read at 405 nm on a kinetic microplate reader (Molecular Device, California). The results presented are the averages of the duplicates, and all experiments were repeated at least three times.
Enzymatic treatment of brain homogenates. Each brain homogenate was treated with 50 μg/ml of proteinase K (Sigma, Missouri) at 37°C for 1 h. The protease was inactivated by the addition of PMSF to a final concentration of 3 mM.
SDS-PAGE and immunoblotting. Similar amounts of recombinant murine, human, bovine, or ovine PrP proteins were dissolved in 2x sample buffer and heated at 95°C for 5 min before separation on 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). For reducing conditions, 2-mercaptoethanol (5% final concentration; Sigma) was added to the sample buffer. The polyacrylamide gel was transferred onto nitrocellulose membrane and probed by MAb 8H4 or MAb 7A12. After incubation with horseradish peroxidase-conjugated goat anti-mouse immunoglobulin G Fc (Chemicon, California), the transferred PrP species were visualized using the chemiluminescence blotting system (Roche Diagnostics, New Jersey).
Sucrose gradient fractionation. Twenty percent total brain homogenate was prepared as described above, and Sarkosyl was added to a final concentration of 1%. After incubation for 30 min on ice, 0.5 ml of homogenate was loaded on a 10-to-60% step sucrose gradient in ultraclear centrifuge tubes (13 by 51 mm). Ultracentrifugation was carried out in SW55 rotors (Beckman, California) at 200,000 x g, 4°C, for 60 min. Fractions of 0.42 ml were collected from the top of the tube. In some experiments, blue dextran (molecular mass, 2,000 kDa; Sigma, Missouri) was included as a marker. To detect PrP species present in different sucrose gradient fractions by immunoblotting, 10 μl of each fraction was mixed with an equal volume of 2x SDS-PAGE buffer and heated at 95°C for 5 min before being loaded onto a 12% gel, and immunoblot assays were carried out as described earlier. For aggregation-specific ELISA, 5 μl of each fraction in 95 μl of PBS was added to the ELISA plate using protocols described earlier.
RESULTS
Rationale for the development of an aggregation-specific ELISA. In a conventional sandwich or capture ELISA, two MAbs with distinct binding epitopes are required: one MAb is immobilized on a solid phase to capture the antigen, and a second MAb that reacts with a distinct epitope is then used to detect the bound antigen. We reasoned that when PrP protein dimerizes or aggregates, some MAb binding epitopes would be buried, while other MAb binding epitopes might be present more than once (see Fig. 5, below). Accordingly, the number of epitopes available for binding will depend on the composition of the aggregate. If the PrP aggregate is a tetramer, then some MAb binding epitopes may be represented four times. We further postulated that if the epitope were present more than once, we might be able to use the same MAb for capture as well as detecting MAb. In addition, since dimers and aggregates are physically larger with repeating epitopes, they will have a stronger avidity with higher probability of being captured by the immobilized antibody, compared to a smaller PrP monomer. Therefore, in our dimer or aggregation-specific ELISA, one MAb is used as the capture MAb as well as the detecting MAb.
Whether all recombinant PrP (rPrP) proteins contain dimer is unresolved (14, 18, 29). We first confirmed that PrP dimer is present in rHu-PrP, rMo-PrP, rOv-PrP, and rBo-PrP proteins by immunoblotting. Using anti-PrP MAb 7A12 for detection, rHu-PrP, rMo-PrP, and rOv-PrP migrate as a 24- to 25-kDa protein with a small amount of dimeric PrP migrating as a 50-kDa protein (Fig. 1A, left panel). On the other hand, rBo-PrP migrates slower due to the presence of an additional octapeptide repeat. All dimeric rPrP contains a disulfide bond, because under reducing conditions only the 24- to 25-kDa monomeric rPrP is detected (Fig. 1A, right panel). Based on densitometry of the bands, the amount of dimeric rPrP is usually less than 5%. MAb 7A12 did not detect any rPrP with a molecular mass larger than a dimer.
Subsequently, we determined whether any of our 30 anti-PrP MAbs could react specifically with rMo-PrP dimer. The locales of some of the MAb binding epitopes are diagrammatically presented in Fig. 1B. Some MAbs react with linear sequences, while others react with conformational epitopes. Only five MAbs, 11G5 (amino acids [aa] 114 to 130), 7A12 (aa 143 to 155), 12H7 (conformational epitope from aa 145 to 200), 7C11, and 8C6 (conformational epitopes from aa 145 to 200) consistently reacted strongly with rMo-PrP (Fig. 1C). Interestingly, neither the N-terminal-specific MAb 8B4 (aa 37 to 44) nor the C-terminal-specific MAb 8F9 (aa 220 to 231) detected any dimeric rMo-PrP; therefore, both the N terminus and the C terminus are not available for binding in the rMo-PrP dimers.
Four of these MAbs, 11G5, 7A12, 12H7, and 7C11, also reacted with rBo-PrP (Fig. 1C), rOv-PrP (Fig. 1C), and rHu-PrP (Fig. 1C) to various degrees. These results suggest that these recombinant PrP dimers share certain common features that are most noticeable in the central region, which contains the MAb 11G5 (aa 114 to 130) and 7A12 (aa 143 to 155) binding epitopes. However, the availability of some of these antibody binding epitopes is species specific. For example, MAb 8C6 reacted preferentially with rMo-PrP, MAb 12A4 reacted only with rBo-PrP, MAb 8F9 reacted most strongly with rOv-PrP, and MAb 5C3 reacted more strongly with rHu-PrP. Overall, the immunoreactivity with rOv-PrP was consistently higher.
The aggregation-specific ELISA is specific for dimeric PrP. Aggregation of rMo-PrP is age dependent. Freshly prepared rMo-PrP does not contain PrP dimer (29). We prepared rMo-PrP proteins that are free of detectable dimeric PrP and rMo-PrP that contains dimer as described previously (3, 29). The presence of dimeric rMo-PrP in the preparation was first confirmed by immunoblotting with MAb 7A12. It was clear that dimeric rMo-PrP was present only in sample A (Fig. 2A) and not in sample B (Fig. 2B). We also used a conventional capture ELISA using MAb 8B4 as the immobilized, capture MAb and biotinylated MAb 7A12 as the detecting MAb to verify that the two preparations of rMo-PrP proteins had comparable amounts of protein (Fig. 2C and D).
We next determined whether the MAbs identified earlier indeed react only with dimeric rMo-PrP. When tested in the dimer-specific ELISA, MAbs 7A12 and 11G5 but not 8H4 or 8B4 reacted strongly with the rMo-PrP preparation, sample A that contains dimeric rMo-PrP, in an antigen concentration-dependent manner (Fig. 2C). On the other hand, none of the tested MAbs (11G5, 7A12, 8H4, or 8B4) was able to react with sample B, which lacks rMo-PrP dimer (Fig. 2D). Furthermore, the dimer-specific ELISA is about 100-fold more sensitive than the immunoblot assays. These results provide strong evidence that our aggregation-specific ELISA indeed reacts with rMo-PrP dimers.
Detection of PrP aggregates in PrPSc-infected mouse brains. We next determined whether the aggregation-specific ELISA could detect PrP aggregates present in brain homogenates from mice infected with the ME7 strain of PrPSc. We screened the 30 anti-PrP MAbs and compared the immunoreactivities detected in infected brains with normal sham-infected control brains. Brain homogenates from Prnp–/–-129/Ola mice were also used as negative controls. To assist the comparison, the results from rMo-PrP are also presented (Fig. 3A). None of the anti-PrP MAbs had significant immunoreactivity with brain homogenates from either sham-infected control mice or Prnp–/–-129/Ola mice (not shown). Only 4 of the 30 anti-PrP MAbs reacted strongly with brain homogenates from infected mice (Fig. 3B).
Interestingly, MAbs 8C6, 7A12, and 12H7, which reacted strongly with rMo-PrP protein, did not react with infected brain homogenates at all. On the other hand, MAbs 7H6, 6H3, and 8F9, which did not react with rMo-PrP, reacted robustly with infected brain homogenates. Only MAb 11G5 reacted with both rMo-PrP proteins and infected brain homogenates. Furthermore, the immunoreactivity differences between infected and normal brain homogenates were profound. For example, when tested with MAb 11G5, the difference in immunoreactivity (as defined by the optical density [OD]) between infected and normal brain homogenate was more than 300%. The immunoreactivity detected with MAb 11G5 was also much stronger in infected brains than in the rMo-PrP preparation. Overall, these results suggest that the nature of the PrP aggregates formed in infected brain and rMo-PrP dimer are different.
By using MAb 11G5, we found that the aggregation-specific ELISA could detect signals in an infected brain homogenate, which contains between approximately 0.6 and 6 μg of total brain proteins (Fig. 3C, left panel). Furthermore, the binding of biotinylated 11G5 was blocked by unconjugated MAb 11G5 in a MAb-specific and concentration-dependent manner. The irrelevant MAb 8B4 did not block (Fig. 3C, right panel). When ELISA plates were first coated with an irrelevant, non-anti-PrP MAb, there was no binding of biotinylated MAb 11G5 (not shown). Therefore, the aggregation-specific ELISA is both antigen and antibody specific.
The aggregation-specific ELISA is applicable to two other strains of PrPSc. We next determined whether our findings in ME7-infected brains are applicable to two other strains of PrPSc, 22L and 139A. Individual brain homogenate was prepared from ME7-infected (n = 4), 22L-infected (n = 4), and 139A-infected (n = 4) mice at the terminal stages of disease, and each homogenate was analyzed individually. We found that MAb 11G5/11G5 also reacted with PrP aggregates in mice infected with either 22L or 139A PrPSc (Table 1). Similar to ME7-infected brains, the immunoreactivity was robust and highly reproducible. Thus, the aggregate-specific assay is applicable to at least three mouse PrPSc strains.
The immunoreactivity detected by the aggregation-specific ELISA is associated with PK-resistant PrP species. One cardinal feature of PrPSc is its proteinase K (PK) resistance (2). We determined whether the binding activity detected with MAb 11G5/11G5 is associated with PK-resistant PrP species in mice infected with ME7 (n = 4), 139A (n = 4), or 22L (n = 4) PrPSc. Each brain homogenate was divided into two tubes. One was treated with PBS, and the other was treated with 50 μg/ml of PK as described earlier. PK digestion did not reduce the binding in 22L-, 139A-, or ME7-infected brains (Table 1). Therefore, the PrP aggregates detected in this assay are PK resistant and most likely represent PrPSc.
The aggregation-specific ELISA reacts with PrPSc aggregates of various sizes in infected brains. Recently, we found that after ultracentrifugation in a 10-to-60% sucrose gradient, all the mouse PrPC species are present in the top fractions, mainly in fractions 1 and 2. In sharp contrast, in PrPSc-infected mouse brain, immunoreactivity is present in all fractions, with the strongest reactivity present in the bottom fractions, fractions 10 and 11 (21). These bottom fractions are known to contain the largest PrPSc aggregates (35).
We fractionated one control brain homogenate and one ME7-infected brain homogenate in a sucrose gradient. Each fraction was then collected and divided into two samples; one sample was immunoblotted with MAb 8H4, which detects all PrP species, to document the distribution of the PrP species, and the second sample was used to detect PrP aggregate using MAb 11G5 in the aggregation-specific ELISA. Consistent with our earlier findings, in control mice, all PrPC binding activity was present in fractions 1 and 2 (Fig. 4A). In contrast, the PrP immunoreactivity in infected brain was dispersed among all fractions, with fractions 10 and 11 having the highest intensities (Fig. 4B).
We did not detect significant MAb 11G5 immunoreactivity in any of the fractions from normal control mice (Fig. 4C). In contrast, MAb 11G5 detected high levels of immunoreactivity in fractions 3, 4, and 5. On the other hand, fractions 10 and 11, which have the largest PrP aggregates, also contained MAb 11G5 immunoreactivity, but the levels were much lower (Fig. 4C). Blue dextran has a molecular mass of about 2,000 kDa, and it partitions in fraction 3 when centrifuged under identical conditions. Therefore, a majority of the PrPSc aggregates detected by this assay are PrP aggregates of heterogeneous size with molecular mass ranging from around 2,000 kDa to larger than 2,000 kDa, but these aggregates are smaller than the largest PrPSc aggregates present in fractions 10 and 11. This experiment has been repeated with three additional ME7-infected brain homogenates as well as brain homogenates from 139A-infected or 22L-infected mice with comparable results (not shown).
Detection of PrP aggregates during disease progression. ME7-inoculated CD-1 mice begin to show signs at about 130 to 160 days postinfection and die within 3 weeks. Previously, we found that PK-resistant PrP species are only detected in animals infected 140 days earlier (21). To determine when the PrP aggregates first become detectable, brain tissues were obtained from sham-infected mice (n = 4), mice infected 30 days (n = 4) or 70 days earlier (n = 4) and not exhibiting clinical signs, mice infected 140 days earlier (n = 4) and exhibiting obvious clinical signs, and mice at a terminal stage (about 170 days) of disease (n = 4). Significant PrP aggregate immunoreactivity was detected in animals 70 days postinfection, at a time when PK-resistant PrP species are undetectable by immunoblotting (21, 22) (Table 2). These results suggest PrP aggregates are detectable in infected animals much earlier than the manifestation of clinical signs and the detection of PK-resistant PrP species.
DISCUSSION
A panel of 30 different MAbs were developed against recombinant PrP (21, 40), and by screening these anti-PrP MAbs, we have identified five MAbs that preferentially react with rMo-PrP dimers in a dimer-specific ELISA (Fig. 5). Most noteworthy are MAbs 11G5 (aa 114 to 130) and 7A12 (aa 143 to 155), which also react strongly with rBo-PrP, rOv-PrP, and rHu-PrP. Therefore, the epitopes recognized by these two MAbs are conserved across these four species. On the other hand, the binding of the other three MAbs is more variable between rPrP from different species. We also identified MAbs that are species specific, which may reflect the conformational differences among the recombinant PrP or PrP dimers from these four animal species.
None of the 30 MAbs has significant binding with brain homogenates from control, sham-infected mice or Prnp–/– mice. Therefore, if dimeric PrPC is present in normal brain, it is not detected with these MAbs. Of the five MAbs that reacted with rMo-PrP dimers, only MAb 11G5 reacted strongly with brain homogenates from ME7-infected mice. On the other hand, MAb 7A12, which reacts strongly with all four tested recombinant PrP dimers, did not react with infected mouse brain homogenates. This result suggests that as a consequence of PrPC-to-PrPSc conversion the helix 1 region (aa 143 to 157) of the molecule has changed. The helix 1 region may be important in the pathogenesis of prion disease. In a cell model, helix 1 of PrP is a major determinant of PrP folding. Disruption of helix 1 prevents the attachment of the glycophosphatidylinositol anchor and the formation of the N-linked glycans (37). In the absence of the glycophosphatidylinositol anchor, helix 1 induces the formation of unglycosylated and partially protease-resistant PrP aggregates. In all sequenced PrPC with the exception of rodent PrPC, this region also contains the sequence DYEDRYYREN, which is composed entirely of hydrophilic amino acids. In rodents PrPC, the second amino acid, tyrosine (Y), is replaced with a tryptophan. It has been suggested that this region is important in the formation of the hydrophilic core and seeding of PrP aggregates (19). This region also contains the YYR epitope, which has been reported to be exposed only in PrPSc and is not available for binding in PrPC (23). Biophysical studies have also provided strong evidence that PrPC-to-PrPSc conversion involves the conversion of -helix 1 to a ?-sheet structure (6, 30, 32). Another explanation for the inability of 7A12 to detect native PrPSc aggregate may be due to the presence of N-linked glycans. The presence of N-linked glycan in PrPSc aggregates may interfere with the binding of MAb 7A12.
On the other hand, MAb 11G5 reacts with both rPrP dimers and PrPSc aggregates in infected brains. The epitope of MAb 11G5 (aa 114 to 130) includes the first ?-strand (aa 128 to 131). Hence, the conformation of this region may be similar between rPrP dimer and PrPSc aggregates, and PrPC-to-PrPSc conversion may not change the overall conformation of this region. In vitro studies using synthetic peptide have also identified residues 119 to 136 on PrPC to be important in the conversion process (12). Analysis of 27 mammalian and 9 avian PrP proteins revealed that the most conserved region outside the globular domain is located between residues 113 and 137; thus, this part of the molecule must be important in the biology of PrPC (38).
Worthy of note is that the binding of MAb 11G5 to infected brain homogenate is about 300% higher than to rMo-PrP. Since rMo-PrP dimer has two MAb 11G5-reactive epitopes, the PrP aggregates present in infected brain may contain multiple PrP molecules with more than two MAb 11G5-reactive epitopes. Furthermore, MAb 11G5 also reacts with brain homogenates from animals infected with either one of the two other strains of mouse PrPSc, namely, 139A and 22L. Hence, while different strains of PrPSc are known to have different conformations, the MAb 11G5 epitope on PrPSc aggregate is shared between three different strains of mouse PrPSc.
We also identified three MAbs, 7H6, 6H3, and 8F9, which did not react with rMo-PrP dimer but reacted significantly with infected brain homogenates. The MAb 7H6-reactive epitope (aa 130 to 140) is contiguous to the MAb 11G5-reactive epitope and right before the helix 1 region. MAb 6H3 reacts with a conformational epitope which is located at the C-terminal region (15). The MAb 6H3-reactive epitope is quite unusual, as its availability for binding is critically dependent on the N terminus (15). Previously, we reported that binding of MAb 6H3 to recombinant rHu-PrP could be blocked by the binding of MAb 8B4, which binds to the N-terminal end of rHu-PrP. Accordingly, we speculated that there might be interactions between the N terminus and C terminus of the rHu-PrP protein. However, the relationship between these observations and the presence of multiple MAb 6H3-reactive epitopes in PrPSc aggregates in infected brains is not clear.
MAb 8F9 (aa 220 to 231) does not react with rMo-PrP dimer in our dimer-specific ELISA but reacts significantly with the PrPSc aggregates in infected brains. Recently, a MAb was generated by immunizing mice with a linear sequence encompassing residues from 214 to 226 of PrP. This MAb reacts with a conformational epitope which is available for binding in PrPSc but not in PrPC or recombinant PrP (8). Both of these results suggest that the conformation of the C terminus is amenable to change during the conversion process.
By ultracentrifugation in a sucrose gradient, most of the immunoreactivity detected by the aggregation ELISA is present in fractions 3, 4, and 5. These fractions are mostly devoid of PrPC, as it partitions to the top two fractions. Interestingly, while the bottom fractions also have immunoreactivity, the levels of binding are much lower in these largest PrPSc aggregates, which indicates that this aggregate-specific ELISA is more efficient in detecting smaller aggregates than the larger one. When fractionated in the same gradient, the 2,000-kDa blue dextran is present in fraction 3. Therefore, the aggregated PrPSc being detected is likely to have a mass similar to blue dextran and larger.
At the present time, the relationship between the PrP aggregates detected by the aggregation-specific ELISA and the largest PrPSc aggregates in the infected brains is not known. Using an in vitro conversion assay, it is estimated that the "seed" or converting activity associated with PrPSc is heterogeneous in size but larger than the molecular mass standard, blue dextran (5). It should be noted that while all fractions from sham-infected mice are sensitive to PK digestion, PK-resistant PrP species are detected in all fractions from infected mice (results not shown). These results suggest that the size of PK-resistant PrP species is heterogeneous. Hence, the aggregation-specific ELISA detects only a subpopulation of the PK-resistant species. The composition of these PrPSc aggregates is not known. In addition to PrP, it is possible that they may contain other cellular components, such as nucleic acids, lipids, non-PrP proteins, polysaccharides, or glycosaminoglycans.
Our time-course studies in ME7-infected mice revealed that the appearance of the PK-resistant PrP species and clinical signs all occur around the same time, at 140 days postinoculation. On the other hand, using MAb 11G5 the accumulation of the PrPSc aggregate was detected earlier, at 70 days postinfection. We did not detect any immunoreactivity in animals infected 30 days earlier. These results are in agreement with our recent finding that by using an epitope-scanning assay to detect changes in the conformations of PrPSc, the earliest change detected was in animals infected approximately 70 days earlier (21). We also found that, at this time, aberrant full-length and truncated PrP species begin to accumulate in the infected brains (22). The PrPSc we are detecting with the aggregation ELISA is PK resistant. The reason that we can detect PrPSc at 70 days postinoculation is most likely because the ELISA is more sensitive than immunoblot assays. The assay can routinely detect significant binding in lysates containing between 0.6 and 6 μg of total brain proteins. Using rMo-PrP as standard, we estimated that the PrPC in normal brain accounts for approximately 0.01% of the total brain proteins (results not shown). Therefore, the aggregation ELISA has the potential to detect between 0.06 and 0.006 μg of aggregated PrP. However, it will require highly purified PrPSc aggregates to precisely determine the sensitivity of the aggregation-specific ELISA. So far, the accumulation of PrP aggregates during disease progression has only been carried out with MAb 11G5. MAb 11G5 is specific for an epitope (aa 115 to 130) at the central region, and this region is exposed in the recombinant PrP dimeric structure (14). The majority of the screened anti-PrP MAbs cannot react with the PrPSc aggregates at the terminal stage of disease, which may be caused by the masking of their binding sites during the progressive aggregation of PrPSc. Therefore, it is possible that by using other MAbs we may be able to detect PrP aggregates at earlier time points after infection, when PrP aggregates are smaller.
Currently, all the in vitro diagnostic tests for prion diseases require either the demonstration of PK-resistant PrP species in brain homogenate or the uncovering of hidden epitopes after GdHCl treatment (30a). The aggregation-specific ELISA described here provides certain advantages: (i) the assay is much more sensitive because of the use of a capture antibody; (ii) the assay is simpler because a denaturing step is not required; and (iii) with the use of more than one antibody, a built-in control can be used to monitor the specificity and sensitivity of the assay. While all the tests were carried out in infected mice, based on our studies with rOv-PrP, rBo-PrP, and rHu-PrP it is likely that the same approach can be used to develop a test for other animal prion diseases as well as in human prion diseases. Indeed, recent results suggest that the aggregate-specific assay also detects PrPSc in Creutzfeldt-Jacob disease patients (B. Chang et al., unpublished data). Finally, the principle of the aggregate-specific ELISA we have developed may be applicable to other diseases caused by abnormal protein aggregation, such as Alzheimer's disease or Parkinson's disease.
ACKNOWLEDGMENTS
This work is supported in part by NIH grants NS-045981-01 (to M.S.S.), AG20245 (to T.W.), and an award/contract from the U.S. Department of Army, DAMD17-03-1-0286 (to M.S.S.) and from BBSRC (8/BSD17730) to (I.M.).
We thank Witold K. Surewicz for providing the recombinant human PrP and Neil Greenspan for suggestions.
REFERENCES
Bellinger-Kawahara, C. G., E. Kempner, D. Groth, R. Gabizon, and S. B. Prusiner. 1988. Scrapie prion liposomes and rods exhibit target sizes of 55,000 Da. Virology 164:537-541.
Bolton, D. C., M. P. McKinley, and S. B. Prusiner. 1982. Identification of a protein that purifies with the scrapie prion. Science 218:1309-1311.
Brown, D. R., I. K. Iordanova, B. S. Wong, C. Venien-Bryan, F. Hafiz, L. L. Glasssmith, M. S. Sy, P. Gambetti, I. M. Jones, C. Clive, and S. J. Haswell. 2000. Functional and structural differences between the prion protein from two alleles prnpa and prnpb of mouse. Eur. J. Biochem. 267:2452-2459.
Caughey, B., and G. S. Baron. 2002. Factors affecting interactions between prion protein isoforms. Biochem. Soc. Trans. 30:565-569.
Caughey, B., D. A. Kocisko, G. J. Raymond, and P. T. Lansbury, Jr. 1995. Aggregates of scrapie-associated prion protein induce the cell-free conversion of protease-sensitive prion protein to the protease-resistant state. Chem. Biol. 2:807-817.
Cohen, F. E. 1999. Protein misfolding and prion diseases. J. Mol. Biol. 293:313-320.
Cohen, F. E., and S. B. Prusiner. 1998. Pathologic conformations of prion proteins. Annu. Rev. Biochem. 67:793-819.
Curin, S. V., M. Bresjanac, M. Popovic, H. K. Pretnar, V. Galvani, R. Rupreht, M. Cernilec, T. Vranac, I. Hafner, and R. Jerala. 2004. Monoclonal antibody against a peptide of human prion protein discriminates between Creutzfeldt-Jacob's disease-affected and normal brain tissue. J. Biol. Chem. 279:3794-3798.
Enari, M., E. Flechsig, and C. Weissmann. 2001. Scrapie prion protein accumulation by scrapie-infected neuroblastoma cells abrogated by exposure to a prion protein antibody. Proc. Natl. Acad. Sci. USA 98:9295-9299.
Griffith, J. S. 1967. Self-replication and scrapie. Nature 215:1043-1044.
Heppner, F. L., C. Musahl, I. Arrighi, M. A. Klein, T. Rulicke, B. Oesch, R. M. Zinkernagel, U. Kalinke, and A. Aguzzi. 2001. Prevention of scrapie pathogenesis by transgenic expression of anti-prion protein antibodies. Science 294:178-182.
Horiuchi, M., G. S. Baron, L. W. Xiong, and B. Caughey. 2001. Inhibition of interactions and interconversions of prion protein isoforms by peptide fragments from the C-terminal folded domain. J. Biol. Chem. 276:15489-15497.
Huang, Z., S. B. Prusiner, and F. E. Cohen. 1996. Structures of prion proteins and conformational models for prion diseases. Curr. Top. Microbiol. Immunol. 207:49-67.
Knaus, K. J., M. Morillas, W. Swietnicki, M. Malone, W. K. Surewicz, and V. C. Yee. 2001. Crystal structure of the human prion protein reveals a mechanism for oligomerization. Nat. Struct. Biol. 8:770-774.
Li, R., T. Liu, B.-S. Wong, T. Pan, M. Morillas, W. Swietnicki, K. O'Rourke, P. Gambetti, W. K. Surewicz, and M.-S. Sy. 2000. Identification of an epitope in the C terminus of normal prion protein whose expression is modulated by binding events in the N-terminus. J. Mol. Biol. 301:567-574.
Lu, B. Y., and J. Y. Chang. 2001. Isolation of isoforms of mouse prion protein with PrPSC-like structural properties. Biochemistry 40:13390-13396.
McKinley, M. P., M. B. Braunfeld, C. G. Bellinger, and S. B. Prusiner. 1986. Molecular characteristics of prion rods purified from scrapie-infected hamster brains. J. Infect. Dis. 154:110-120.
Meyer, R. K., A. Lustig, B. Oesch, R. Fatzer, A. Zurbriggen, and M. Vandevelde. 2000. A monomer-dimer equilibrium of a cellular prion protein (PrPC) not observed with recombinant PrP. J. Biol. Chem. 275:38081-38087.
Morrissey, M. P., and E. I. Shakhnovich. 1999. Evidence for the role of PrPC helix 1 in the hydrophilic seeding of prion aggregates. Proc. Natl. Acad. Sci. USA 96:11293-11298.
Pan, K. M., M. Baldwin, J. Nguyen, M. Gasset, A. Serban, D. Groth, I. Mehlhorn, Z. Huang, R. J. Fletterick, F. E. Cohen, and S. B. Prusiner. 1993. Conversion of alpha-helices into beta-sheets features in the formation of the scrapie prion proteins. Proc. Natl. Acad. Sci. USA 90:10962-10966.
Pan, T., R. Li, S. C. Kang, B. S. Wong, T. Wisniewski, and M. S. Sy. 2004. Epitope scanning reveals gain and loss of strain specific antibody binding epitopes associated with the conversion of normal cellular prion to scrapie prion. J. Neurochem. 90:1205-1217.
Pan, T., P. Wong, B. Chang, C. Li, R. Li, S. C. Kang, T. Wisniewski, and M. S. Sy. 2005. Biochemical fingerprints of prion infection: accumulations of aberrant full-length and N-terminally truncated PrP species are common features in mouse prion disease. J. Virol. 79:934-943.
Paramithiotis, E., M. Pinard, T. Lawton, S. LaBoissiere, V. L. Leathers, W. Q. Zou, L. A. Estey, J. Lamontagne, M. T. Lehto, L. H. Kondejewski, G. P. Francoeur, M. Papadopoulos, A. Haghighat, S. J. Spatz, M. Head, R. Will, J. Ironside, K. O'Rourke, Q. Tonelli, H. C. Ledebur, A. Chakrabartty, and N. R. Cashman. 2003. A prion protein epitope selective for the pathologically misfolded conformation. Nat. Med. 9:893-899.
Peretz, D., R. A. Williamson, Y. Matsunaga, H. Serban, C. Pinilla, R. B. Bastidas, R. Rozenshteyn, T. L. James, R. A. Houghten, F. E. Cohen, S. B. Prusiner, and D. R. Burton. 1997. A conformational transition at the N terminus of the prion protein features in formation of the scrapie isoform. J. Mol. Biol. 273:614-622.
Priola, S. A., B. Caughey, K. Wehrly, and B. Chesebro. 1995. A 60-kDa prion protein (PrP) with properties of both the normal and scrapie-associated forms of PrP. J. Biol. Chem. 270:3299-3305.
Prusiner, S. B. 1982. Novel proteinaceous infectious particles cause scrapie. Science 216:136-144.
Prusiner, S. B. 1998. Prions. Proc. Natl. Acad. Sci. USA 95:13363-13383.
Prusiner, S. B., M. P. McKinley, K. A. Bowman, D. C. Bolton, P. E. Bendheim, D. F. Groth, and G. G. Glenner. 1983. Scrapie prions aggregate to form amyloid-like birefringent rods. Cell 35:349-358.
Qin, K., D. S. Yang, Y. Yang, M. A. Chishti, L. J. Meng, H. A. Kretzschmar, C. M. Yip, P. E. Fraser, and D. Westaway. 2000. Copper(II)-induced conformational changes and protease resistance in recombinant and cellular PrP. Effect of protein age and deamidation. J. Biol. Chem. 275:19121.
Safar, J., P. P. Roller, D. C. Gajdusek, and C. J. Gibbs, Jr. 1993. Conformational transitions, dissociation, and unfolding of scrapie amyloid (prion) protein. J. Biol. Chem. 268:20276-20284.
Safar, J., H. Wille, V. Itri, D. Groth, H. Serban, M. Torchia, F. E. Cohen, and S. B. Prusiner. 1998. Eight prion strains have PrP(Sc) molecules with different conformations. Nat. Med. 4:1157-1165.
Sigurdsson, E. M., M. S. Sy, R. Li, H. Scholtzova, R. J. Kascsak, R. Kascsak, R. Carp, H. C. Meeker, B. Frangione, and T. Wisniewski. 2003. Anti-prion antibodies for prophylaxis following prion exposure in mice. Neurosci. Lett. 336:185-187.
Speare, J. O., T. S. Rush III, M. E. Bloom, and B. Caughey. 2003. The role of helix 1 aspartates and salt bridges in the stability and conversion of prion protein. J. Biol. Chem. 278:12522-12529.
Swietnicki, W., R. Petersen, P. Gambetti, and W. K. Surewicz. 1997. pH-dependent stability and conformation of the recombinant human prion protein PrP(90-231). J. Biol. Chem. 272:27517-27520.
Tompa, P., G. E. Tusnady, P. Friedrich, and I. Simon. 2002. The role of dimerization in prion replication. Biophys. J. 82:1711-1718.
Tzaban, S., G. Friedlander, O. Schonberger, L. Horonchik, Y. Yedidia, G. Shaked, R. Gabizon, and A. Taraboulos. 2002. Protease-sensitive scrapie prion protein in aggregates of heterogeneous sizes. Biochemistry 41:12868-12875.
Wille, H., G. F. Zhang, M. A. Baldwin, F. E. Cohen, and S. B. Prusiner. 1996. Separation of scrapie prion infectivity from PrP amyloid polymers. J. Mol. Biol. 259:608-621.
Winklhofer, K. F., J. Heske, U. Heller, A. Reintjes, W. Muranyi, I. Moarefi, and J. Tatzelt. 2003. Determinants of the in vivo folding of the prion protein. A bipartite function of helix 1 in folding and aggregation. J. Biol. Chem. 278:14961-14970.
Wopfner, F., G. Weidenhofer, R. Schneider, A. von Brunn, S. Gilch, T. F. Schwarz, T. Werner, and H. M. Schatzl. 1999. Analysis of 27 mammalian and 9 avian PrPs reveals high conservation of flexible regions of the prion protein. J. Mol. Biol. 289:1163-1178.
Yin, S. M., Y. Zheng, and P. Tien. 2003. On-column purification and refolding of recombinant bovine prion protein: using its octarepeat sequences as a natural affinity tag. Protein Expr. Purif. 32:104-109.
Zanusso, G., D. Liu, S. Ferrari, I. Hegyi, X. Yin, A. Aguzzi, S. Hornemann, S. Liemann, R. Glockshuber, J. C. Manson, P. Brown, R. B. Petersen, P. Gambetti, and M. S. Sy. 1998. Prion protein expression in different species: analysis with a panel of new mAbs. Proc. Natl. Acad. Sci. USA 95:8812-8816.(Tao Pan, Binggong Chang, )