Applying the tools of chemistry (mass spectrometry and covalent modification by small molecule reagents) to the detection of prions and the study of their structure

Abstract

Prions are molecular pathogens, able to convert a normal cellular prion protein (PrP^C) into a prion (PrP^Sc). The information necessary for this conversion is contained in the conformation of PrP^Sc. Mass spectrometry (MS) and small-molecule covalent reactions have been used to study prions. Mass spectrometry has been used to detect and quantitate prions in the attomole range (10⁻¹⁸ mole). MS-based analysis showed that both possess identical amino acid sequences, one disulfide bond, a GPI anchor, asparagine-linked sugar antennae, and unoxidized methionines. Mass spectrometry has been used to define elements of the secondary and tertiary structure of wild-type PrP^Sc and GPI-anchorless PrP^Sc. It has also been used to study the quaternary structure of the PrP^Sc multimer. Small molecule reagents react differently with the same lysine in the PrP^C conformation than in the PrP^Sc conformation. Such differences can be detected by Western blot using mAbs with lysine-containing epitopes, such as 3F4 and 6D11. This permits the detection of PrP^Sc without the need for proteinase K pretreatment and can be used to distinguish among prion strains. These results illustrate how two important chemical tools, mass spectrometry and covalent modification by small molecules, are being applied to the detection and structural study of prions. Furthermore these tools are or can be applied to the study of the other protein misfolding diseases such as Alzheimer Disease, Parkinson Disease, or ALS.

Mass Spectrometry-Based Study of the Structure of PrP^Sc

Mass spectrometry has played a significant role in determining the primary covalent structures of PrP^C and PrP^Sc. A combination of genetic analysis,¹ amino acid sequencing,^2-5 chemical,⁶ and MS analysis^3,7-9 was used to show that PrP^C and PrP^Sc possess identical amino acid sequences, a single disulfide bond, two sites for the attachment of sugar antennae, and a covalently bound glycosylphosphatidylinisotol (GPI) anchor (Fig. 1). A detailed mass spectrometry-based comparison of the sugar antennae that were present in purified PrP^C and PrP^Sc showed that the antennae of both isoforms were highly varied (>30 different forms) and that they varied similarly in both PrP^C and PrP^Sc.^10–12 Mass spectrometric analysis of the GPI anchor showed that several variants were found in similar proportions in both PrP^C and PrP^Sc. ^{13, 14} These results indicated that any post-translational differences between PrP^C and PrP^Sc were not in the primary structure and were not covalent.

Figure 1. Cartoon of the hamster PrP^C. The location of the epitopes of the AG4, 3F4, R1, and 6D11 monoclonal antibodies and tryptic peptides GENFTETDIK, VVEQMCTTQYQK, ESQAYYDGR, and YPGQGSPGGNR and the chymotryptic peptide YRPVDQY are indicated. The position of Met 213 is indicated M.

A variety of mass spectrometry-based approaches have been used to study the chemistry and structure of PrP. The location of methionines oxidized by hydrogen peroxide in rPrP was determined by mass spectrometry.¹⁵ More recently, mass spectrometry was used to study the structure of rPrP and rPrPβ by measuring the difference in the hydrogen peroxide reactivity of methionines present in the two isoforms.¹⁶ Copper-mediated oxidation of the histidine residues in PrP was studied by mass spectrometry-based analysis of rPrP.¹⁷ The reaction products of lysines and reactive oxygen species in hamster PrP^Sc were quantitated by mass spectrometry.¹⁸ The differences in secondary structure of the product from the seeded polymerization of rPrP starting with PrP^Sc or rPrP were examined using hydrogen/deuterium (H/D) exchange and MS analysis.¹⁹ Mass spectrometry-based analysis of the reaction of cross-linking reagents and PrP^Sc has been used to study the tertiary structure of hamster PrP^Sc by measuring the spatial proximity of terminal glycines in hamster PrP 27–30.²⁰ Other researchers used different cross-linking reagents and PK digestion to study the structural differences between rPrP and rPrPβ.²¹ These approaches used recombinant PrP or regions of PrP^Sc that were not glycosylated.

Mass spectrometry-based study of PrP^Sc structure has mainly focused on the non-glycosylated N-terminal region of PrP^Sc. Mass spectrometry has been used to study copper-mediated oxidative cleavage of PrP^C.²² MS was used to identify N-terminal ragged ends and prion strain-dependent variations in those N-terminal ragged ends.^23-26 The PK cleavage sites between cysteine 179 and the N-terminus of the Dy and 263K strains of PrP^Sc were identified by a MS-based analysis.²⁷ This analysis showed that the PK sensitive and PK resistant forms of PrP^Sc were infectious and shared a common structure.^28,29 Due to the glycoform variability, these approaches have been limited to the N-terminal region of the protein.

The PrP produced by transgenic GPI-anchorless (GPI^-) mice was shown to have no GPI anchor and only a limited amount of glycosylation.^30,31 Such limited glycosylation permitted a more complete MS analysis of H/D exchange that occurred in this region. It showed only a limited amount of exchange, which suggested that GPI^- PrP^Sc was composed mostly of β-sheet secondary structure.³² Other researchers used mass spectrometry to identify the products from PK digested GPI^- PrP^Sc.^33,34 This analysis led the researchers to conclude that PrP^Sc was composed of largely PK resistant β-sheet strands that were linked by identifiable short, flexible, and PK sensitive loops. Mass spectrometry was used to detect variations in the extent of H/D exchange as rPrP was converted into infectious rPrP^Sc.³⁵ Recent MS-related work has led to the reassessment of the relative proportion of α-helix and β-sheet in PrP^Sc samples, whose purity was assessed by mass spectrometry.³⁶ These studies concluded that the secondary structure of PrP^Sc was composed almost entirely of β-sheet.

Mass Spectrometry-Based Analysis of Oxidized PrP^Sc

During the MS analysis of the primary structure of hamster PrP^Sc, one of the peptides (IMERVVEQMCTTQYQK; Hamster PrP 205–220) was observed to contain oxidized methionine. The origin of this modification was unclear, since it could have been the consequence of a biological process or an artifact.³ When methionines in human rPrP were systematically replaced by a more polar analog, methoxinine, the methoxinine containing rPrP was found to be more susceptible to fibrilization than the native human rPrP.³⁷ Using different model systems, other researchers obtained results which suggested that the oxidation of methionines in this region of rPrP would make it more susceptible to fibrilization.^38,39 Western blot-based analysis of PrP^Sc showed that methionines from this region were oxidized to a significant extent.^40-42 Oxidized methionine was proposed to be a covalent signature of PrP^Sc, which would be the first demonstrated covalent difference between PrP^C and PrP^Sc.⁴⁰

A large body of work suggested that oxidized methionines were unlikely to have a role in prion propagation. Mammals were demonstrated to express methionine sulfoxide reductases (Msr) as a housekeeping protein whose purpose was to reduce any oxidized methionines.^43-46 The incubation period for prion infected Msr-ablated mice was shown to be the same as that of the corresponding wild type mice, which suggested that oxidized methionines do not have a role in prion propagation.⁴⁷ Peptides spanning this region of PrP were shown to fibrilize without the presence of oxidized methionine.^3,48 Other studies showed that the presence of oxidized methionines actually inhibited PrP fibrilization.^49-51 When rPrP (29-231) was oxidized by hydrogen peroxide, enzymatically cleaved, and then analyzed by MS, the methionines present in the peptide IMERVVEQMCTTQYQK were shown to be remarkably resistant to hydrogen peroxide-mediated oxidation,¹⁵ which was consistent with predictions from a more general study of hydrogen peroxide-mediated oxidation of methionines.⁵² These results suggested that significant biological oxidation of this region of PrP^Sc was unlikely to occur.

Since a methionine could be oxidized by a biological process or by artifact, it was important to have a method that minimized artifactual oxidation when quantitating methionine oxidation. SDS-PAGE separation was essential for a Western-blot analysis of PrP, but it was also shown to be an inherently oxidative process.^53,54 Although methionine oxidation from SDS-PAGE could be minimized by rigorous use of chemical reductants,⁵³ this approach was not used in the analysis of PrP by Western blot analysis. Methionines were shown to be oxidized by exposure to air during general handling^55-57 and even, to a limited extent, by the ionization necessary for MS analysis.^58,59 These reports indicated that artifactual oxidation of methionine caused by common laboratory manipulation could be significant.

MS-based analysis was determined to be well suited for the detection and quantitation of the amounts of oxidized methionine present in PrP.^60,61 Solution-based digestion was used instead of a SDS-PAGE-based digestion to eliminate that source of artifactual oxidation.^62,63 The well established multiple reaction monitoring (MRM) and solution-based digestion methods were used to determine the ratio of the oxidized and unoxidized forms of the tryptic peptide VVEQMCTTQYQK (Hamster PrP 209–220).^62-66 The levels of artifactual oxidation as a result of in-solution digestion and MS analysis was determined to be detectable, but negligible.^60,61 This work indicated that a MRM-based analysis introduced a minimal amount of artifactual methionine oxidation.

A MRM-based approach was used to analyze three strains of hamster-adapted scrapie, ten field cases of sheep scrapie and ten cases of elk CWD.^60,61 The prions were purified according to the method of Bolton et al.,⁶⁷ which isolated both the PK sensitive and resistant fractions of PrP^Sc in high yield.⁶⁸ The results from a 10-week time course experiment showed that the proportion of oxidized methionine in the analyte peptide actually decreased from the earliest time point (3 wk; ~13%) to the final time point (10 wk, ~8%) (Fig. 2).⁶¹ Analysis of brains from hamsters infected with the 263K, 139H or drowsy (Dy) prion strain showed that the proportion of oxidized methionine present in PrP^C (uninfected hamsters) was similar (~2–10%) to that found in PrP^Sc derived from hamsters infected with those strains of hamster-adapted scrapie.⁶¹ Samples from ten field cases of scrapie-infected sheep and ten field cases of CWD-infected elk showed a similarly low proportion of methionine oxidation in the homologous peptide (VVEQMCITQYQR; Sheep PrP 212–223) in both the sheep and elk PrP^Sc.⁶⁰ Even though the sheep and elk were much older than the hamsters and presumably contained PrP^Sc that had more time to be exposed to biological oxidation, the proportion of oxidized methionine was similar in sheep, elk, and hamsters.^60,61 These data indicated that the observed amounts of oxidation present in these methionines was a consequence of handling proteins in the presence of air and not evidence of a PrP^Sc specific covalent signature.^60,61

Figure 2. Percentage of oxidized Met213 present in PrP27–30 isolated from the brains of hamsters during the course of an ic challenge (263K), as measure by mass spectrometry. Results are reported as means (± SD) for each time point (n = 4).

The relationship between polymorphisms in the VVEQMCITQYQR peptide and the extent of methionine oxidation was determined using a model system.^60,69 Unnatural sheep rPrP polymorphisms (I replaced by T [hamster analog] or V [mouse analog]) were isolated, digested, mixed and subjected to air oxidation.^60,69 MRM analysis of the peptide mixtures showed that peptides containing isoleucine were oxidized in a higher proportion than in analogs containing valine or threonine (I > V > T), even though all three were exposed to air oxidation under the same conditions. These results showed that the chemistry of air oxidation was different from that of peroxide mediated oxidation. Furthermore, it indicated that sheep and elk PrP^C were more susceptible to oxidation than was hamster PrP^C. Even though sheep and elk PrP were intrinsically more susceptible to oxidation, sheep, elk and hamster PrP^Sc all had similarly low levels of oxidized methionine. This body of work provided further evidence that post-translational differences between PrP^C and PrP^Sc were purely conformational and not covalent.

Mass Spectrometry-Based Detection of PrP^Sc

The MRM approach was used to detect and quantitate the PrP^Sc present in sheep, elk, deer, mouse and hamsters.^{60,62,63,70,71} The set of tryptic peptides from the digestion of rPrP was analyzed and the peptides VVEQMCTTQYQK (Hamster PrP 209–220), VVEQMCITQYQR (Sheep PrP 212–223), and VVEQMCVTQYQK (Mouse PrP 208–219) were empirically determined to be suitable for a MRM-based analysis.^62,71 These analyte peptides were detectable in the attomole (10⁻¹⁸ mole) range and present in both PrP^Sc and PrP 27–30. MS analysis of tissue from uninfected controls showed that they contained no peptides that would interfere with the MRM analysis. Stable isotope-labeled (¹⁵N and ¹³C) analogs of these analyte peptides were chemically synthesized and used as internal standards.^60-63,71 Adding a known amount of an internal standard to the tryptic digest of a sample permitted the quantitation of the analyte peptide relative to the known amount of the added internal standard. In hamsters, this approach was used to detect PrP^Sc one day post ic inoculation (Fig. 3).⁶³ In addition PrP^Sc was easily detectable in non-obex brain tissue from field cases of sheep scrapie and elk CWD.^60,71 Furthermore this approach was used to detect and quantitate PrP^Sc in non-brain tissue (spleen, tonsil and RAMALT).

Figure 3. Scheme showing the process of analyzing a prion sample by a MS-based MRM method. An aliquot (~1/5) of a hamster brain (1 d post ic inoculation) was processed for mass spectrometry (Steps 1–7; +PK). The sample was chromatographed using a nano-LC system and then continuously sprayed by electrospray ionization (ESI) (Step 8) into the mass spectrometer. The first quadrupole (Q1) was set to permit only the ions with a mass/charge ratio (m/z; z = 2) of the analyte peptide (VVEQMCTTQYQK) to enter the collision cell (q2). In the collision cell the filtered ions were fragmented. These fragments entered the third quadrupole which was set to permit ions with an m/z corresponding to an optimized fragment of the analyze peptide (b₂ ion [VV]) to enter the detector. The resulting signal from the detector was recorded.

The tryptic peptides GENFTETDIK (Hamster PrP 195–204) and ESQAYYDGR (Hamster PrP 221–229) were used to confirm the diagnosis of prion diseases.^62,70,71 They were not suitable for use as analyte peptides, since the asparagine in the GENFTETDIK peptide was only present in those PrP molecules that were not glycosylated (N-197) and ESQAYYDGR was not present in all forms of PrP^Sc.⁷² These two peptides were shown to have a greater MRM signal intensity than the analyte peptide, so they could be used to confirm the presence of PrP^Sc without a loss of sensitivity.^62,71 This approach was used to detect prions in five strains of hamster-adapted scrapie and four strains of murine-adapted scrapie in asymptomatic preclinical animals.^70,71 Furthermore, it was used to confirm the presence of prions in field cases of sheep infected with scrapie and elk infected with CWD.^60,71 These results demonstrate the utility of mass spectrometry to diagnose prion diseases.^70,71

MRM-based methods of detecting PrP were not limited to three tryptic peptides. The peptides YPGQGSPGGNR (Hamster PrP 38–48) and YRPVDQY (Hamster PrP 163–169) were found to be common to virtually all known mammalian PrP proteins and could be used to detect PrP, using rPrP as a model.^73,74 Both of these peptides were detected in digested PrP^Sc from the brains of prion-infected hamsters (Silva CJ, unpublished results). Mass spectrometry-based analysis of N-terminal cleavage sites (ragged ends) was used to distinguish between different types of human prion diseases.²³ The MS-based analysis of the ragged ends from sheep PrP^Sc showed utility in distinguishing among strains of sheep scrapie.^24-26 The variety of MS-based prion detection methods was illustrated by these reports.

Mass spectrometry has been employed to analyze complex intact protein, proteomic, and peptidomic mixtures and to detect differences between prion-infected animals and healthy controls. The urine of BSE-infected cattle was analyzed by this method and shown to be subtly different from uninfected control animals.^75-77 A proteomic analysis of glycosylated serum proteins from prion infected mice showed differences in the relative amounts of these proteins.⁷⁸ Other researchers used a whole protein approach to show that there were differences in the protein expression profiles present in the brains of prion infected mice and uninfected controls.⁷⁹ A more focused approach used MS to examine differences in the hippocampal proteome in control mice and those in the late stage of an infection (Me7 prion strain).⁸⁰ An analogous whole protein comparison of serum from scrapie-infected sheep showed differential protein expression profiles in infected animals examined at early and later stages of the disease.⁸¹ A peptidomics-based approach has been used to distinguish between prion infected and healthy animals by examining the peptidome of hamsters infected with prions.⁸² A MS-based peptidomic analysis was able to distinguish differences in the relative amounts of tryptic peptides derived from the small proteins (<10 kDa) present in the CSF fluid from CJD and non-CJD patients.⁸³ A proteomics-based approach revealed that urine-derived fertility products were contaminated with PrP.^84,85 Proteomic approaches have been used to identify proteins that co-purify with PrP^Sc as part of a search for PrP^Sc binding partners.^86-88 Mass spectrometry has been used to determine the binding partners of GPI^- PrP.⁸⁹ Other researchers used mass spectrometry to identify peptides from bovine brain tissue that accelerate conformational changes in rPrP.⁹⁰ These approaches reveal the myriad ways in which mass spectrometry has been used to detect prions directly or to infer their presence indirectly.

Small Molecule Based Study of PrP^Sc

Mass spectrometry-based and other forms of analysis have clearly demonstrated that the post-translational differences between PrP^C and PrP^Sc are not covalent and must be conformational (non-covalent).⁹ These results meant that the same amino acid could react differently with the same reagent, depending on whether it was in the PrP^C or PrP^Sc conformation. A variety of reagents have been used to covalently modify PrP, but not to distinguish between PrP^C and PrP^Sc.^{20,21,91–95} These results indicated that there are amino acid residues on PrP^Sc that could be covalently modified by a variety of reagents and these covalent bonds would remain intact after PrP^Sc was denatured. Thus, some information about the PrP^Sc conformation would be retained after the reacted PrP^Sc was denatured.

Purified PrP^Sc has been reacted with commercially available reagents to obtain structural information about its C-terminal region.⁹⁶ Hamster PrP^Sc was reacted with tetranitromethane (TNM) or acetic anhydride (Ac₂O), which nitrated the tyrosine or acetylated the ϵ-amino groups of lysine, respectively. After reaction, the samples were analyzed by mass spectrometry to determine which tyrosine or lysine was covalently modified. The extent of tyrosine nitration was also monitored by Western blot, using the antibody R1 (Hamster PrP 225–231 epitope).⁹⁷ Covalent modification of the tyrosine interfered with the binding of the antibody, resulting in a reduction of signal relative to the unmodified tyrosine. In this way antibodies were used to detect covalent modifications of PrP^Sc.⁹⁶

The lysines present in the PrP^C conformer reacted completely with synthetic reagents. The extent of this reaction was monitored by Western blot using antibodies that did and did not have lysine-containing epitopes.⁹⁸ The NHS esters of acetic acid (Ac-NHS) and 4-trimethylamoniumbutyric acid ([3-carboxypropyl]trimethylammonium chloride) (tMAB-NHS) were used and covalently converted the ϵ-amino group of an accessible lysine into the corresponding amide. These reagents were reacted with detergent solubilized hamster brain homogenates (DSHBH) from uninfected animals and analyzed by Western blot using the antibodies 3F4 (Hamster PrP 109–112 epitope) and AG4 (Hamster PrP 31–51 epitope) (Fig. 4). The epitope of the 3F4 antibody was shown to contain a lysine, while the epitope of the AG4 antibody was shown not to (Fig. 1). Blots probed with the 3F4 antibody showed no signals for PrP^C after reaction with either of these reagents. In contrast blots probed with AG4 showed signals from PrP^C after reaction with both of the reagents. These results indicated that covalent modification of a lysine that was part of an antibody’s epitope (3F4) would prevent the binding of that antibody. They also indicated that non lysine-containing epitopes (AG4) were not affected by the reagents.

Figure 4. Western blots of uninfected or 263K-infected hamster DSHBH. Samples consist of uninfected or 263K-infected DSHBH reacted with nothing (- reagent), 20 mM tMAB-NHS (+tMAB), or 20 mM Ac-NHS (+Ac). Identical amounts of the same reaction mixtures were probed with either the 3F4 or AG4 monoclonal antibody.

These reagents were used to distinguish between the PrP^C and PrP^Sc present in prion infected DSHBH.⁹⁸ The Ac-NHS and tMAB-NHS reagents were reacted with DSHBH from animals infected with the 263K strain of hamster-adapted scrapie. The reaction mixtures were analyzed by Western blot and probed with either AG4 or 3F4. The blot probed with AG4 was qualitatively similar to the blot of the uninfected DSHBH (Fig. 4B). When the same blot was probed with 3F4 signals were apparent in both the unreacted and reacted DSHBH from infected animals, but not from uninfected animals. Since the 3F4 epitope was hidden in the PrP^Sc conformation, it was less able to react with the reagents.⁹⁹ When the reacted PrP^Sc was denatured for Western blot analysis, the now exposed and unreacted epitope could bind to the 3F4 antibody and would generate the observed signal. Since the 3F4 epitope was exposed in the PrP^C conformation, this approach could be used to detect PrP^Sc in the presence of PrP^C without the need to remove the PrP^C first.

The Ac-NHS and the tMAB-NHS reagents both reacted with lysines, but to a different extent that was influenced by the non-NHS portions of their chemical structures. These differences in chemical structure meant that they interacted differently with the chemical environment of the lysine prior to reacting with it. Such reagent-dependent differences in reactivity could be seen by Western blot analysis (Fig. 4A; lanes 4–6), where the signal for the Ac-NHS reaction was noticeably less intense than that of the tMAB-NHS reagent. The tMAB-NHS reagent was positively charged and the Ac-NHS reagent was not, so such differences were not surprising. Differences in the extent of reactivity were also observed when the non-NHS portion of the reagents used were bulkier, but uncharged.¹⁰⁰ This indicated that the extent of the reaction was dependent upon the choice of reagents.

Since these reagents could be used to distinguish between the PrP^C and PrP^Sc conformations, they might be useful in distinguishing among PrP^Sc conformers or strains.⁹⁸ The drowsy (Dy) and 263K strains were reacted with the Ac-NHS, tMAB-NHS, and other related reagents and analyzed by Western blot. The blots were probed with 3F4 or 6D11 (Hamster PrP 98–101 epitope), another antibody shown to have a lysine-containing epitope. The blots showed that the reactivity of the Dy conformation was different from that of the 263K conformation.^98,100 This showed that these two lysines (3F4 or 6D11 epitope) are in different chemical environments depending upon their PrP^Sc conformation (Dy or 263K).

Only a limited number of PrP lysines were demonstrated to be a component of an available antibody’s epitope. Covalent modification of lysine would also prevent its cleavage by trypsin, so a measurement of the absence of tryptic peptides was determined to be a means to detect covalently modified lysines. The absence of tryptic peptides was measured by MS using stable isotope-labeled internal standards. This approach has been used to detect covalently modified lysines in recombinant hamster PrP.^100,101 Such a MS–based approach could be used to analyze PrP^Sc lysines until appropriate antibodies become available.

Mass Spectrometry and Small Molecule-Based Study of Other Protein Misfolding Diseases

Other, non-TSE, neurodegenerative diseases have been associated with pathological protein misfolding and discussed in recent reviews.^102,103 These include the amyloid-β deposits associated with Alzheimer, the accumulation of tau associated with Alzheimer disease or other tauopathies, α-synuclein associated with Parkinson disease, and the SOD1 associated with amyotrophic lateral sclerosis (ALS). The proteins associated with these diseases show protein misfolding and progressive pathological aggregation that are analogous to those seen in prion diseases. As with prions, researchers have used small molecule modification and mass spectrometry to study these proteinopathies.

Some recent reports showed that mass spectrometry could be used to study the structure of these misfolded proteins and to detect them in tissues. The structure of α-synuclein oligomers and a pathogenic variant of SOD1 have been studied using mass spectrometry to monitor differences in hydrogen/deuterium exchange.^104,105 MS has been used to study the aggregation and post-translational modifications of SOD1.¹⁰⁶ It was used to identify novel α-synuclein isoforms in brain tissue and amyloid-β in cerebral spinal fluid.^107-109 These results demonstrated the utility of MS for structural studies.

These proteins have been shown to have varied post-translational modifications, which have been detected and characterized by mass spectrometry. The increase in nitration of tyrosine-39 of α-synuclein was monitored by mass spectrometry.¹¹⁰ Phosphorylation sites of tau and other proteins were identified in the brains of Alzheimer disease patients using a novel mass spectrometry-based approach.¹¹¹ Components of the Lewy body proteome were determined using mass spectrometry.¹¹² A mass spectrometry-based analysis has been used to study the structure α-synuclein by analyzing the reaction products of small molecule crosslinkers with α-synuclein.¹¹³ Antibodies have been used to detect covalent modification of these proteins. Nitrated forms of α-synuclein associated with Parkinson disease have been identified using antibodies specific for the nitrated form of the protein.^114,115 These results indicate that MS or antibodies can be used to detect the specific post-translational modifications associated with these proteinopathies.

In principle the covalent modification by small molecules approach used to study prions (vide supra) could be applied to the study of amyloid-β, tau, α-synuclein, or SOD1. The reagents have been described and a number of monoclonal antibodies (mAbs) have been produced. Anti-amyloid-β (1–40) monoclonal antibodies were shown to bind to a lysine-containing region of the protein.¹¹⁶ Several anti-tau mAbs, with lysine-containing eptiopes, have been produced.^117,118 The anti-α-synuclein mAb Syn-17 was demonstrated to have a lysine-containing epitope.¹¹⁹ Some of the lysine residues of SOD1 were shown to be components of the epitopes of anti-SOD1 monoclonal antibodies.¹²⁰ As has been shown with prions, these or other antibodies (with lysine-containing epitopes) may be useful in developing a Western blot-based detection of conformation dependent differences in small molecule reactivity between the normal and pathogenic forms of amyloid-β, tau, α-synuclein, or SOD1.

Summary

Mass spectrometry has played a crucial role in the study of prions and other protein misfolding diseases and will continue to do so. Mass spectrometry-based analysis provides the strongest evidence for conformation being the sole difference between PrP^Sc and PrP^C. More recently mass spectrometry has been used to study the structure of PrP^Sc, by analyzing the peptide mixtures resulting from proteolytic digestion. The widespread availability of GPI^- PrP^Sc and rPrP^Sc means a ready supply of non-glycosylated PrP protein that is suitable for MS analysis.

Using antibodies to detect small molecule modifications of PrP^Sc is another means of studying PrP^Sc. This approach does not require a mass spectrometer, the reagents are reasonably easy to prepare, and it uses equipment that is available in any molecular biology laboratory. It relies on the well established technique of Western blotting. This approach can be used to distinguish between PrP^C, PrP^Sc, and different strains of PrP^Sc. Furthermore it can be used to study other protein misfolding diseases (Alzheimer’s disease, Parkinson’s disease, etc.). In principle it can be used instead of MS analysis when the appropriate antibodies are available.

Two important chemical tools, mass spectrometry and covalent modification by small molecules, are being successfully applied to the detection and structural study of these molecular pathogens.

10.4161/pri.27891

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.