Prion protein conversion in vitro
- First Online:
- Cite this article as:
- Supattapone, S. J Mol Med (2004) 82: 348. doi:10.1007/s00109-004-0534-3
The infectious agents of prion diseases are composed primarily of an infectious protein designated PrPSc. In cells infected with prions, a host glycoprotein termed PrPC undergoes induced conformational change to PrPSc, but the molecular mechanism underlying this structural transition occurs remains unknown. The prion-seeded conversion of PrPC to protease-resistant PrPSc-like molecules (PrPres) has been studied both in crude and purified in vitro systems in order to investigate the mechanism of protein conformational change in prion disease. Conversion of purified PrPC into PrPres is specific with respect to species-dependent and polymorphic differences in PrP sequence as well as biophysical variations between prion strains, recapitulating the specificity of prion propagation in vitro. The protein misfolding cyclic amplification (PMCA) technique, which utilizes crude brain homogenates, produces much higher yields of PrPres than conversion of purified PrP molecules, suggesting that additional cellular factors may stimulate PrPres formation. In a modified version of the PMCA technique, PrPres from diluted prion-infected brain homogenate can be amplified > ten-fold when mixed with normal brain homogenate without sonication or the anionic detergent sodium dodecyl sulfate (SDS). Under these conditions, PrPres amplification in vitro depends upon both time and temperature, has a neutral pH optimum, and does not require divalent cations. In vitro PrPres amplification is inhibited by both reversible and irreversible thiol blockers, indicating that the conformational change from PrPC to PrPres requires a thiol-containing factor. Stoichiometric transformation of PrPC to PrPres in vitro also requires specific RNA molecules, suggesting that host-encoded catalytic RNA molecules may play a role in the pathogenesis of prion disease. Heparan sulfate stimulates conversion of purified PrPC into PrPres in vitro, and heparan sulfate proteoglycan molecules are required for efficient PrPres formation in prion-infected cells. Future studies using in vitro PrPres conversion and amplification assays promise to provide new mechanistic insights about the PrP conversion process, and to generate clinically useful tools.
Prion diseases such as kuru, variant Creutzfeldt Jakob Disease (vCJD), bovine spongiform encephalopathy (BSE), chronic wasting disease (CWD), and scrapie are fatal neurodegenerative diseases that can be transmitted naturally or experimentally to normal hosts. The unconventional infectious agents responsible for transmission, which are highly resistant to inactivation and lack informational nucleic acids, have been termed prions . Prions consist primarily of a misfolded protein designated PrPSc, and replicate by inducing a host-encoded glycoprotein designated PrPC to undergo a conformational change to form new PrPSc molecules in a self-propagating process .
In contrast to PrPC, which is composed of 42% α-helix and 3% β-sheet, PrPSc contains 30% α-helix and 43% β-sheet [3, 4, 5]. As a result, PrPSc is substantially more detergent-insoluble and protease-resistant than PrPC, allowing the two isoforms to be differentiated biochemically. Purified PrP molecules in either conformation contain a C-terminal glycosylphosphatidylinositol anchor , two N-linked oligosaccharides attached to residues N181 and N197, and a single intramolecular disulfide bridge between residues C179-C214 .
The precise mechanism by which infectious PrPSc proteins induce host PrPC molecules to undergo conformational change and create new PrPSc molecules is currently undetermined. Most notably, it is unknown whether any molecules other than PrPSc and PrPC are required to produce new prions in vivo. The structural and chemical dynamics of the conversion process have also remained obscure. For instance, it has not been determined whether the intramolecular disulfide bridge in PrPC remains intact during conformational change to PrPSc, or whether this bond must break and reform to permit structural rearrangement of the protein. To investigate various aspects of the mechanism of PrP conversion biochemically, several groups have used cell-free systems in which PrPC molecules are converted into protease-resistant, PrPSc-like molecules (PrPres) as a result of their interaction with pre-existing PrPSc molecules.
Comparison of in vitro PrP conversion techniques
Cell-free conversion technique
In vitro PrPres amplification technique
Purified, radiolabeled PrPC derived from tissue culture cells
Crude, post-nuclear supernatant or synaptic plasma membranes of non-infected brain homogenates
Crude, post-nuclear supernatant of prion-infected brain homogenates
Stoichiometry (molar ratio of PrPSc to PrPC)
% PrPC conversion
Saborio and Soto developed a method termed protein misfolding cyclic amplification (PMCA) that efficiently amplified PrPres in hamster brain homogenates through repeated cycles of direct sonication in the presence of the anionic detergent sodium dodecyl sulfate (SDS) . In contrast to PrPres conversion with purified PrP molecules, PMCA is performed with a proportional excess of input PrPC molecules compared to PrPSc, and produces stoichiometric amplification of PrPres during the procedure. Whereas cell-free conversion of purified PrP molecules requires a 50-fold molar ratio of PrPSc to PrPC to drive the formation of PrPres, PMCA can be performed with molar ratios of PrPSc to PrPC less than 1:100.
Basic parameters of PrPres amplification
The observation that in vitro PrP conversion is more efficient in crude brain homogenates than in purified preparations suggested that additional unidentified host factors may be required for efficient PrP conversion. Therefore, a modified version of the PMCA technique was used to investigate the mechanism of PrPres formation biochemically. The first step was to determine whether PrPres could be amplified in vitro without sonication or SDS, because both of these factors have the capability of denaturing biological macromolecules and have also been shown previously to affect PrP structure [18, 19, 20]. The following discussion focuses primarily upon findings obtained using this modified in vitro PrPres amplification assay, which is technically simple and rapid to perform.
In vitro PrPres amplification was optimal at pH values between 6 and 8, depending on the animal species analyzed . These determinations suggest that the pathogenic conformational change from PrPC to PrPres would occur most efficiently in cellular locations with nearly neutral pH environments, such as on the extracellular surface of the plasma membrane  or within the cytoplasm . The optimal pH range of PrPres amplification is similar to the previously determined optimal pH range of PrP conversion in mixtures of lipid raft preparations containing PrPC and brain microsomes containing PrPSc .
Experiments utilizing the chelators EDTA and EGTA showed that divalent ions are not required to generate PrPres from PrPC, suggesting that they do not participate directly in the conversion process in vitro . However, this observation does not rule out an important role for divalent cations in determining PrPres conformation. Other investigators have found that, in some experiments, copper assists refolding of denatured PrPres into infectious prions . Also, chelation of divalent cations alters strain-specific biochemical characteristics of PrPres molecules in CJD-infected brains . Thus, binding of metal ions such as copper and manganese to PrPres may influence its conformation and generate different prion strains.
Kinetic and temperature control experiments showed that PrPres is amplified in a time-dependent manner at 25°C and 37°C, but not at 4°C. After incubation at 37°C for 48 h, PrPres is amplified ~15-fold compared to input PrPres. Notably, the rate of PrPres amplification appears to decrease over time; at 37°C, the initial doubling time of ~2.5 h increases to ~24 h after 2 days . This decrease in PrPres amplification rate over time suggests that at least one of the required amplification components is either depleted or inactivated during the incubation period.
Specificity of PrPres amplification
Transmission barriers limit the efficient propagation of infectious prions between different animal species [26, 27]. The molecular basis for species-specific transmission barriers appears to be differences in the amino acid sequences of PrP molecules between the inoculum and host [28, 29]. To investigate whether the efficiency of PrPres amplification in vitro correlates with species susceptibility to prion infection in vivo, PrPres amplification was compared in three different hamster species whose PrP amino acid sequences mismatch at six positions. Syrian, Armenian, and Chinese hamsters have different scrapie incubation times when inoculated intracerebrally with Syrian hamster-derived Sc237 prions. Whereas Syrian hamsters had a scrapie incubation time of ~69 days, Armenian hamsters had an incubation time of ~174 days, and Chinese hamsters had an incubation time ~ 344 days . When brain homogenates were prepared from each hamster species and separately mixed with Sc237 scrapie brain homogenate, an inverse correlation between PrPres amplification and scrapie incubation time was observed . The rank order of Sc237 PrPres amplification efficiency was Syrian>Armenian>Chinese hamsters, and therefore the degree of PrPres amplification in vitro correlates with the degree of species susceptibility to the infectious prion in vivo.
A single animal species can host several different prion strains that produce distinct clinical and pathological phenotypes [31, 32]. The ability of each strain to produce a unique disease phenotype is preserved upon serial passage. In some cases, PrPres molecules from different strains can also be distinguished biochemically. An example of a strain containing altered PrPres molecules is “drowsy,” which was originally isolated from prion-infected mink and subsequently passaged into hamsters [33, 34, 35]. Drowsy prions contain PrPres molecules that migrate ~2 kDa faster than other hamster prion strains following proteinase K digestion, and drowsy PrPres molecules amplified in vitro also exhibit this slight difference in electrophoretic mobility following proteinase K digestion, suggesting that strain-specific biochemical characteristics can be maintained during PrPres amplification .
Thus, the PMCA-based, non-denaturing in vitro PrPres amplification system shares specific features with the pathogenic process of PrPSc formation in vivo, and provides a model of infectious prion propagation that can be studied biochemically. These findings recapitulated the strain- and species-specific in vitro conversion of purified PrP molecules originally described in the cell-free radiolabel assay [9, 10].
Free sulfhydryl groups are required for PrPres amplification
Experiments with selective thiol blocking compounds led to the discovery that free sulfhydryl groups are required for PrPres amplification in vitro. Several lines of evidence showed that thiol blockade functionally inhibits the conversion of PrPC to PrPres in vitro : (1) N-ethyl maleimide (NEM) and two other selective thiol blockers inhibit PrPres amplification in vitro; (2) thiol blockers also inhibit PrPres amplification in a purified membrane preparation; (3) pre-quenching NEM prevents inhibition; and (4) pre-treatment of normal brain homogenate with NEM inhibits subsequent PrPres amplification.
The amino acid sequence of PrP contains two cysteine groups at positions 179 and 214. Preparations of PrPC and PrPres isolated from hamster brain predominantly contain molecules with intramolecular disulfide bridges . It is currently unclear whether the disulfide bond of PrPC must be broken during the process of conformational change to PrPres. Denaturation of purified preparations of PrPres aggregates by guanidinium chloride releases PrP monomers with an intramolecular disulfide bond . To explain this observation, some investigators have suggested that PrP conformational change occurs without reduction of the disulfide bond , and others have hypothesized that PrPres contains intermolecular disulfide bonds, which undergo disulfide-shuffling during denaturation [37, 38, 39]. Reduction of recombinant PrP polypeptides causes these molecules to shift conformation from predominantly α-helical to predominantly β-sheet [40, 41, 42]. Thus, it is also possible that reduction and reformation of the intramolecular disulfide bond occurs during the conformational change from PrPC to PrPres.
In vitro conversion of purified PrPC to PrPres was not inhibited by 50 mM NEM or 70 mM 2-aminoethyl-methane-thiosulfonate (AEMTS) .
Pre-treatment of normal brain homogenate with NEM inhibits PrPres amplification, whereas pre-treatment of scrapie brain homogenate with NEM has no effect on PrPres amplification .
Disulfide-bonded multimers could not be detected in amplified PrPres samples .
Because PrPC molecules do not contain free thiol groups , the remaining possible mechanisms of PrPres formation are consistent with the existence of a novel thiol-containing cofactor required for PrP conversion (Fig. 2, mechanisms 1–3). The eventual identification of the thiol-containing cofactor by biochemical techniques will be of interest both from a mechanistic standpoint and also as a potential therapeutic target.
Specific RNA molecules stimulate PrPres amplification
Additional experiments demonstrated that the RNA cofactors that stimulate PrPres amplification are species specific. Whereas RNA prepared from hamster and mouse brain successfully stimulated PrPres amplification in vitro, RNA prepared from Escherichia coli, Saccharomyces cerevisiae, Caenorhabditis elegans, and Drosophila melanogaster were inactive . This apparent species specificity cannot be attributed to tissue specificity because total hamster liver RNA also stimulated PrPres amplification. These data argue that invertebrates lack functional homologues of the specific RNA molecules required for PrPres amplification in mammalian cells. RNA molecules were also shown to stimulate and be required for efficient PrPres amplification by the PMCA technique. Addition of total hamster brain RNA increased the PrPres signal obtained after eight sonication cycles of PMCA by ~ ten-fold, and more PrPres was detected at every sonication cycle when additional RNA was present .
Taken together, these results indicate that specific cellular RNA molecules are required for and catalyze PrPres amplification in vitro. An immediate practical application is that RNA preparations can be used to increase the sensitivity of diagnostic tests based on PrPres amplification, such as the PMCA method. From this perspective, the next logical step is to purify the specific stimulatory RNA molecules from appropriate sources and enrich them for use in such assays.
It is important to caution that biochemical studies alone cannot confirm that RNA and PrP molecules interact in vivo. Nonetheless, it is useful to consider the hypothesis that specific host RNA molecules might also catalyze PrPres formation in vivo. Consistent with this hypothesis, nucleic acids bind avidly to, and promote conformational change of recombinant PrP [23, 45, 46, 47, 48, 49, 50, 51, 52]. Some inhibitors of PrPres formation in prion-infected cells, e.g. pentosan polysulfate and heparan sulfate, mimic nucleic acid structure and may act as competitive inhibitors of RNA binding ; while other inhibitors, e.g. polypropyleneimine and polyamidoamine dendrimers, potently bind nucleic acids and may sequester RNA from binding to PrP [54, 55]. It is well established that specific RNA molecules can spontaneously fold into structures that catalyze biological reactions, such as intron splicing and protein translation, and ribonucleoprotein complexes regulate a diverse assortment of cellular processes . Notably, the catalytic RNA hypothesis is consistent with the protein-only hypothesis proposed by Prusiner , because the RNA molecules described are host-encoded and not contained within the infectious agent.
How and where might PrP interact with RNA molecules in a living cell? PrP molecules are known to traffic to various cell compartments where RNA is normally localized, such as the cytoplasm and nucleus, [57, 58, 59, 60]. In addition, a fraction of PrP molecules exist in a transmembrane form with a cytoplasmic domain . Another possibility is that RNA molecules might enter the extracellular space as a result of cell death or active transport [62, 63]. Whether RNA participates in PrPSc formation in living cells and animals, as it appears to do in vitro, will be an important avenue for investigation in the future.
Sulfated polysaccharides termed glycosaminoglycans (GAGs) and GAG-containing proteoglycans are another class of molecules that have been shown to stimulate prion conversion in vitro. Neuronal proteoglycans are particularly enriched in heparan sulfate GAGs, and therefore attention has been focused upon heparan sulfate proteoglycan (HSPG) as a candidate cofactor in the prion conversion process. In the cell-free radiolabel conversion assay, both heparan sulfate and pentosan polysulfate stimulate the formation of PrPres from purified PrPC . In reconstitution experiments, addition of purified heparan sulfate increased the infectivity of dimethyl sulfoxide-solubilized prion rods, whereas treatment with heparinase III, a heparan sulfate-degrading enzyme, reduced infectious titer . In cultured prion-infected cells, treatment with heparinase III or estradiol β-d-xyloside, a proteoglycan glycosylation inhibitor, reduced PrPSc levels. Sodium chlorate, an inhibitor of GAG sulfation, also inhibited PrPSc formation in prion-infected cells, and this effect was partially reversed by addition of exogenous heparan sulfate or chondroitin sulfate . HSPG molecules appear to be co-receptors for PrP, and an HSPG-interaction site on PrP has been identified [67, 68]. These various observations from different systems suggest that endogenous HSPG molecules are likely to play an important role in PrPSc formation in vitro and in vivo.
Biochemical studies of in vitro PrPres formation have opened many new avenues for prion research, with the promise of new mechanistic insights and even the generation of clinically useful diagnostic or therapeutic tools. For many years, the central focus in prion research was to identify the chemical nature of the infectious agent, and this area remains an area of active inquiry. For instance, it is unknown whether all prion infectivity is contained within PrPres or whether a subset of infectious prions is composed of conformationally altered, but protease-sensitive PrP molecules . The most critical sequel to the in vitro experiments described here is to determine the infectivity of PrPres that is newly formed in vitro. If infectious prions can be generated in vitro, the technique of PrPres amplification could be used to facilitate a host of detailed biochemical studies of the infectious agent. The level of PrPres formed by mixing purified PrP molecules alone is too low to detect changes in infectious titer. The level of PrPres amplification by PMCA may be high enough to detect an increase in infectivity by animal bioassay, but cycles of sonication in SDS may cause changes in infectious titer unassociated with PrPres formation [18, 19, 20]. The level of PrPres amplification seen when crude brain homogenates are mixed together without sonication or SDS is probably too low to detect a change in infectivity by conventional bioassays, but this might be circumvented by the stimulation with RNA and/or measurement using recently developed infectivity assays with higher accuracy . In addition to measuring infectivity, it will also be important to determine whether newly formed PrPres molecules are auto-catalytic within the amplification reaction. That is, can PrPres molecules formed in vitro themselves induce the formation of additional PrPres molecules? Such autocatalytic behavior would be predicted if PrPres molecules formed in vitro are infectious.
In vitro PrP conversion assays may also be applied to practical projects, which may benefit public health. For example, such assays could be adapted to make a molecular assessment of species susceptibility to specific strains of infectious prions, at greater speed and more safely and more humanely than inoculation experiments. Raymond et al. used purified molecules to evaluate the risk of humans and livestock species to BSE and CWD, and found that the transmission barrier for humans was similar with these two strains . In vitro PrPres amplification using crude brain homogenates could be used to determine whether genetic factors other than PrP sequence, such as species-specific RNA converting factors, might also affect individual susceptibility to these and other infectious prion strains. Two particular problems with CWD are that the disease affects wildlife populations, and the mode of transmission is unknown. Therefore a whole host of animal species including predators, scavengers, and animals sharing ecological niches with cervids are potentially exposed to the infectious agent. Direct inoculation and long term containment of wild animals would be difficult, expensive, unsafe, and inhumane. Therefore, in vitro PrPres amplification assays may provide the only feasible method to assess the potential wildlife reservoir for CWD.
Currently, a rapid and sensitive prion detection technique is urgently needed to help control the spread of CWD in North America as well as BSE and vCJD in Europe. Such a technique could be used to identify newly infected cases, facilitating public health programs aimed at eradicating these diseases. The amplification of PrPres in vitro could in principle be used to increase the sensitivity of tests based on PrPres detection, such as the standard protease digestion-Western blot assay , filter retention assay , ELISA , or conformation dependent assay . It may be possible to adapt in vitro amplification of prion infectivity to increase the sensitivity of tests such as traditional bioassays or cell-based infectivity assays .
Finally, it may eventually be possible to perform PrPres amplification assays in high-thoughput formats, as has recently been achieved for PrP conversion assays using purified PrP molecules . High-thoughput assays would facilitate screening for drug compounds that could be used therapeutically to combat prion diseases, all of which are currently incurable. The PrPres amplification method offers several potential advantages as a screening assay. The technique is capable of identifying compounds that bind directly to PrP or to important conversion cofactors. Also, PrPres amplification can be applied immediately to medically relevant prion strains. Cell culture-based drug screening strategies are limited in this regard because only certain mouse and sheep scrapie strains have been successfully propagated in cultured cells.
In summary, in vitro PrP conversion and PrPres amplification systems have already contributed significant insights into the molecular mechanism of prion disease, including the basis of species and strain specificity, as well as the participation of free sulfhydryl groups and specific RNA molecules. These studies offer an excellent opportunity to increase our understanding of the biophysical properties of infectious PrP molecules and host factors required for PrP conversion. They may also lead to important advances in public health and clinical management of prion diseases.