Engineering of a protein probe with multiple inputs and multiple outputs for evaluation of alpha synuclein aggregation states.

Abstract

The aggregation of α-synuclein (αS) into oligomers and fibrils is implicated in the pathology of Parkinson’s Disease (PD). While a molecular probe for rapid and comprehensive evaluation of αS aggregation states is critical for a better understanding of PD pathology, identification of therapeutic candidates, and the development of early diagnostic strategies, no such probe has yet to be developed. A structurally flexible αS variant, PG65, was previously developed as a target binding-driven, conformation-switching molecular probe for rapid αS oligomer detection. Though informative, detection using PG65 provides no comprehensive assessment of the αS aggregation states. In the present study, we report engineering of a molecular probe, PG65-MIMO (a PG65 variant with Multiple-Inputs and Multiple-Outputs), that rapidly (within 2 hr) produces comprehensive information on αS aggregation states. PG65-MIMO generates distinct fluorescence responses to the three major αS conformers (monomers, oligomers, and fibrils). PG65-MIMO also displays unique fluorescent signals for αS oligomers, depending on the tris(2-carboxyethyl)phosphine (TCEP) concentration. Our results suggest that the TCEP dependent signaling of PG65-MIMO may be associated with its conformational states. Overall, our study illustrates engineering of an αS variant to create a molecular probe for handling multiple inputs and multiple outputs, addressing the technological gap in αS detection.

1. INTRODUCTION

Aggregation of the intrinsically disordered protein, α-synuclein (αS), has been implicated in the pathogenesis of Parkinson’s Disease (PD), the second most common neurodegenerative disorder [1, 2]. PD is characterized by the loss of dopaminergic neurons in the substantia nigra and the prevalence of intraneuronal Lewy Bodies and Lewy Neurites, which are composed of αS aggregates [3, 4]. αS monomers are intrinsically disordered [5] and self-assemble to form soluble αS oligomeric aggregates [6, 7]. αS oligomers can further aggregate into fibrillar species [8]. Although αS fibrils are rich in cross β sheet structure, αS oligomers display distinct β sheet structure, and there has been accumulating evidence that suggests that these αS oligomers are the major toxic species in the pathology of PD [9].

The 140 amino-acid αS protein is comprised of three regions, an amphipathic N terminus (M1-K60), a hydrophobic non-amyloid-β component (NAC) domain (E61-V95) and an acidic carboxyl terminus (K96-A140) (Fig. 1). The NAC domain is aggregation-prone and critical in αS self-assembly [10, 11]. The NAC domain is comprised of regions with high β sheet potential which are connected by linker regions (N65-G68 and E83-G86) with high turn potential [12]. Physicochemical properties of these linker regions, which are dependent on the constituting amino acids, play a significant role in the aggregation of αS [12].

Figure 1.

Amino acid sequences of αS, PG65 and PG65-MIMO. A few representative amino acid numbers of αS are shown. The sequence fragments that differ among αS, PG65 and PG65-MIMO are underlined. The first linker region (NVGG) is colored in blue and the tetra-cysteine motif in red.

By exploiting the structural flexibility of intrinsically disordered αS and the conformation-sensitiveness of a fluorescent dye, an αS variant, PG65, was developed as a target binding-driven, conformation-switching molecular probe for αS oligomer detection [13]. PG65 generates fluorescence signals rapidly (~2 hr) specific for αS oligomers over αS monomers and fibrils [13], serving as a valuable tool that complements currently available techniques to detect αS oligomers [13, 14], such as ELISA [15]. PG65 was created by the mutation of the first linker region (N65-G68) within the NAC domain (Fig. 1). The linker was substituted with the tetra-cysteine motif, CCPGCC (Fig. 1), which serves as the binding site for a conformation-sensitive bi-arsenical dye, FlAsH (fluorescein arsenical hairpin binder). FlAsH becomes strongly fluorescent upon binding to the tetra-cysteine motif via covalent arsenic-thiol binding [16]. FlAsH fluorescence also depends on conformations of a tetra-cysteine motif and its neighboring sequence [16, 17].

While quantitative detection of αS oligomers using PG65 is informative, more comprehensive evaluation of soluble αS aggregation states may offer information on oligomerization potentials. For example, concentration of αS monomers may reflect thermodynamic and kinetic potentials of oligomerization in samples [18]. Unfortunately, detection using the αS oligomer-specific probe, PG65, provides no information on other αS conformers, generating incomplete depiction of αS aggregation states in samples [13]. Moreover, no molecular probe is currently available for the comprehensive yet rapid evaluation of αS aggregation states. Such molecular probes will be valuable in exploring any correlation between αS aggregation profiles and their biological effects, advancing the development of early diagnostic strategies, identification of aggregation inhibitors of therapeutic relevance and to gaining a better understanding of a molecular basis of PD pathology [19].

In this study, we report engineering of a PG65 variant, PG65-MIMO (PG65 variant with Multiple-Inputs and Multiple-Outputs), a molecular probe that generates fluorescence responses rapidly (~ 2hr) in an αS conformer-dependent manner. PG65-MIMO was created by replacing PG65’s N- and C-terminal residues next to the tetra-cysteine motif with HRW and KTF, respectively. PG65-MIMO displayed distinct fluorescence responses to αS monomers, oligomers, and fibrils. Moreover, PG65-MIMO exhibited different signal behaviors for αS oligomers depending on tris(2-carboxyethyl)phosphine (TCEP) concentrations. Our data support a view that the TCEP dependent signaling of PG65-MIMO may be associated with its conformational states. Overall, we postulate that αS conformer-specific, TCEP-dependent signaling by PG65-MIMO can be utilized as a rapid and comprehensive profiling method for αS aggregation states.

2. MATERIALS AND METHODS

2.1. Reagents

Oligonucleotides were purchased from Integrated DNA Technologies, Inc. (Coralville, IA, USA). High-fidelity Platinum Pfx DNA polymerase, Escherichia coli BL21 (DE3) cells, ampicillin, and the amyloid oligomer-specific antibody, A11, were purchased from Thermo Fisher Scientific (Waltham, MA, USA). DNA purification kits were purchased from Qiagen (Valencia, CA, USA). FPLC columns for αS purification and characterization were purchased from GE Healthcare (Rahway, NJ, USA). Restriction enzymes and Instant Sticky-end Ligase Master Mix were purchased from New England Biolab (Ipswich, MA, USA). αS sequence-specific antibodies, F11, 5C2, 211 and D10, were purchased from either Santa Cruz Biotechnology, Inc (Dallas, TX, USA) or Novus International (Saint Charles, MO, USA). FlAsH-EDT₂ was purchased from Santa Cruz Biotechnology, Inc (Dallas, TX, USA). All other reagents were purchased from Thermo Fisher Scientific, unless otherwise specified.

2.2. DNA construction of PG65-MIMO

PG65-MIMO was genetically constructed using the plasmid pRK172 harboring the human αS gene [12]. Briefly, the DNA sequence coding for PG65-MIMO was created by overlap extension PCR using 5’-GGTTTTCACCGTCATCACC-3’ (forward primer), 5’-GCAACAGCCCGGACAGCACCAACGATGCTCTTTGGTCTTCTCAGCCACTGTTGCCAC AC-3’ (reverse primer), 5’-TGCTGTCCGGGCTGTTGCAAAACCTTTACGGGTGTGACAGCAGTAGCCCAGAAG-3’ (forward primer), 5’-CTTTCAGCAAAAAACCCCTCA-3’ (reverse primer) and pRK172 as the template. This was followed by purification via QIAquick PCR purification and QIAquick gel-extraction kits. The DNA product was then digested by NdeI and HindIII restriction enzymes and ligated into the plasmid pRK172, which was digested with the same enzymes. Ligation of the PG65-MIMO gene into the plasmid was completed using Instant Sticky-end Ligase Master Mix. Ligated products were then transformed into 50 μL of E.coli BL21 cells using a Bio-Rad Gene Pulser (Hercules, CA, USA). Electroporated cells were incubated for 1 hr at 250 rpm and 37 °C in a New Brunswick Scientific Innova TM4230 incubator (Edison, NY, USA). Electroporated cells were then plated onto LB agar plates containing 100 μg/mL of ampicillin and subsequently incubated at 37 °C overnight. A single colony was selected the following day and then grown in LB media containing 100 μg/mL ampicillin. The recombined plasmid, pRK172-PG65-MIMO, was extracted, purified from the cells, and sequenced.

2.3. Protein Expression and Purification

The αS expression from the plasmid pRK172 and purification have been previously described [12, 13]. Briefly, 1% of overnight starter culture cells were added to 1 L of LB medium containing 100 μg/mL of ampicillin and grown to an OD₆₀₀ of ~0.6. Expression of αS was induced by the addition of isopropyl-β-D-1-galactopyranoside to a final concentration of 1 mM. The cell culture was then shaken for additional 18–20 hrs at 250 rpm and 25 °C and pelleted by centrifugation at 2,500 g for 15 minutes in a Beckman Coulter Avanti JE centrifuge (Fullerton, CA, USA). Cell pellets were stored at −80°C until purification.

For purification, the cell pellets were defrosted and resuspended in Tris Buffer (50 mM Tris-HCl, pH 8.0). The cells were lysed by sonification using a Hielscher UP200S ultrasonicator (Hielscher Ultrasonics, Teltow, Germany), while on ice for two 5-minute intervals. The cell lysate was centrifuged for 1 hr at 27,000 xg. The supernatant was then collected, and heat treated at 80 °C for 20 minutes to denature endogenous cellular proteins, followed by a second centrifugation at 27,000 xg for 1 hr. αS was further purified from the heat-treated supernatant by FPLC using anion exchange and size exclusion chromatography columns. Sample fractions collected from the FPLC columns were analyzed on an SDS-PAGE gel and the fractions with the highest purity (>95%) were combined. The combined fractions containing αS was then buffer exchanged against dH₂O using a desalting column, aliquoted into microcentrifuge tubes and lyophilized using a Labonco FreeZone Freeze Dryer (Kansas City, MO, USA). Identity of αS was confirmed by SDS-PAGE and dot blot assays using αS sequence-specific antibodies, F11, 5C2, 211 and D10. Lyophilized αS was stored at −80 °C.

PG65-MIMO was expressed from the plasmid pRK172-PG65-MIMO and purified similarly, but with the addition of 1 mM TCEP in all buffers during purification. PG65-MIMO was aliquoted into multiple microcentrifuge tubes and stored at −80 °C. A PG65-MIMO sample was thawed immediately prior to use and never frozen again after thawing.

2.4. αS Monomer, Oligomer and Fibril Preparation

For the preparation of αS monomers, lyophilized αS was solubilized in Phosphate Buffered Saline with Azide (PBSA) (20 mM Na₂HPO₄/NaH₂PO₄, 150 mM NaCl, 0.02% (w/v) NaN₃) and then filtered through an EMD Millipore Amicon Ultra filter (100 kDa MWCO, Millipore Sigma, Burlington, MA, USA). αS monomers were collected in the filtrate. The 100 kDa cutoff filter was used to collect αS monomers (MW: ~14 kDa) because of their natively unfolded nature, making their hydrodynamic size similar to that of ~60 kDa globular proteins [5].

For αS oligomer preparation, lyophilized αS was dissolved in PBSA at 560 μM and filtered through a 0.45 μm syringe filter. Aliquots were taken from the filtered αS solution and its αS concentration was estimated using A₂₈₀ in 8M urea. The αS concentration of the filtered solution was then adjusted to 350 μM by adding buffer prior to transferring to a glass vial. Glass vials containing aqueous αS were subjected to orbital shaking at 250 rpm in a 37 °C incubator for 6 hrs. After the culmination of 6 hrs, the αS solution was filtered with an Amicon Ultra 100 kDa MWCO filter. αS oligomers were collected in the retentate. The retentate was washed 3 times with PBSA during collection and then stored at −80 °C. αS oligomer prepared in this manner seeded αS fibrillization, as shown in our previous studies [13, 20].

For αS fibrils, the filtered αS solution was freshly prepared at 350 μM, similar to αS oligomer preparation, and then transferred to a glass vial with continuous magnetic bar stirring for ~2 to 3 weeks at 37 °C. After the incubation, the samples were centrifuged to collect insoluble αS fibrils. Prior to use, the fibrils were washed repeatedly with PBSA and αS concentrations of the fibril samples were estimated by back calculation of αS that was lost during washing using A_280. The insoluble αS fibrils were resuspended in PBSA prior to use.

Concentrations of αS conformers used in study correspond to their monomer-equivalent molar concentrations, unless otherwise specified.

2.5. Circular Dichroism (CD) Spectroscopy

Secondary structure of αS was analyzed by CD using a Jasco J-815 CD spectrometer. The CD spectra were collected from 205–250 nm with a 0.1 cm pathlength cuvette. The sample spectrum was adjusted by subtraction of the background buffer spectrum and converted to mean residue molar ellipticity (deg·cm²·dmol⁻¹).

2.6. Thioflavin T (ThT) Fluorescence

Five μL of αS samples were mixed with 10 μL of 0.1 mM ThT solution and 195 μL of PBSA. ThT fluorescence was then measured on a Photon Technology QuantaMaster QM-4 spectrofluorometer (Horiba, Kyoto, Japan), with excitation at 440 nm and emission monitored at 485 nm.

2.7. Size Exclusion Chromatography (SEC)

Size exclusion chromatography (SEC) was conducted using a precision packed column, Superdex 200 PC 3.2/300, on the FPLC (Akta Purifier, GE Healthcare). Briefly, 100 μL of samples were injected into the column at a flowrate of 0.1 ml/min and elution peaks were monitored via A₂₈₀.

2.8. Transmission Electron Microscopy (TEM)

Five μL of samples were pipetted onto glow-discharged 400 Mesh copper grids and then negatively stained with 1% uranyl acetate solution in deionized water. The samples were imaged on a FEI Talos L120C Transmission Electron Microscope (FEI Corp., Hillsboro, OR, USA) with a Gatan 4k x 4k OneView camera in the Microscopy Laboratory (RRID: SCR_017934) at NYU Langone Medical Center.

2.9. SDS-PAGE and native-PAGE

SDS-PAGE was used to confirm the size and purity of the protein. For SDS-PAGE, the protein was loaded onto a NuPAGE™ 4–12% Bis-Tris, 1.0 mm, Mini Protein Gel (ThermoFisher Scientific) and ran on a XCell Surelock Mini-Cell gel tank (Thermo Fisher Scientific) with NuPAGE™ MES SDS Running Buffer (Thermo Fisher Scientific).

Native-PAGE with in-gel fluorescence imaging was used to examine binding between PG65-MIMO and labeled αS. For native-PAGE, the protein sample was loaded onto a Native-PAGE™ 4–16% Bis-Tris gel and ran on the XCell Surelock Mini-Cell gel tank with Native-PAGE™ Running Buffer (Thermo Fisher Scientific).

2.10. Intrinsic Tryptophan Fluorescence

Intrinsic tryptophan fluorescence of PG65-MIMO was measured on the Photon Technology QuantaMaster QM-4 spectrofluorometer. The excitation wavelength was set at 295 nm and the fluorescence emission intensity measured at 350 nm. PG65-MIMO at 0.5 μM in PBSA, was titrated incrementally with TCEP during intrinsic tryptophan fluorescence measurements.

2.11. Measurement of FlAsH Fluorescence

Fresh aliquots of 150 μM PG65-MIMO was diluted to 0.33 μM in PBSA containing EDT and TCEP in the presence and absence of αS. The mixtures were incubated at 25 °C for 1 hr in siliconized microtubes, followed by the addition of FlAsH (in the form of bis EDT adduct, FlAsH-EDT₂) to a final concentration of 1.5 μM FlAsH. The FlAsH concentration of 1.5 μM was chosen because FlAsH fluorescence of PG65 (0.33 μM) was saturated at ≥ ~ 1 μM of FlAsH with and without αS (data not shown). The final concentration of EDT was 100 μM. The final concentration of TCEP was either 75 μM or 150 μM, unless otherwise mentioned. The samples were incubated for an additional 1 hr to allow sufficient time for FlAsH to bind to PG65-MIMO. The mixtures were then transferred onto a Corning® 96 well NBS™ black microplate (Corning, NY, USA) and fluorescence measurements were taken on a Biotek™ Synergy™ H1 microplate reader (Winooski, VT, USA). The samples were excited at 500 nm and the fluorescence emission intensity was measured at 534 nm. Other than fluorescence emission intensity changes, there was no notable shift of the emission spectra at 528–550 nm of PG65-MIMO in the presence and absence of αS monomers, oligomers and fibrils. Overall, the time required for FlAsH fluorescence-based αS detection using PG65-MIMO was 2 hr, beyond which no significant additional FlAsH fluorescence signals were generated (data not shown). When measured with 1.5 μM of FlAsH, background FlAsH fluorescence of αS monomers was ~100 times lower and αS fibrils was ~500 times lower than FlAsH fluorescence of PG65-MIMO at the same protein concentrations, indicating negligible non-specific FlAsH binding to these αS species. The background FlAsH fluorescence of αS oligomers was ~1/20 of FlAsH fluorescence of PG65-MIMO, indicating the greater degree of non-specific FlAsH binding to oligomeric αS, relative to the other αS conformers. To take into account the background FlAsH fluorescence of αS, relative FlAsH fluorescence changes (RFCs) of PG65-MIMO were calculated, according to the following equation: RFC = [FI (PG65-MIMO + αS) – FI (PG65-MIMO) – FI (αS)] / FI (PG65-MIMO); where FI (PG65-MIMO + αS), FI (PG65-MIMO) and FI (αS) = fluorescence emission intensity at 534 nm of the mixture containing PG65-MIMO and αS, PG65-MIMO alone and αS alone, respectively.

2.12. In-gel Fluorescence Imaging

αS monomer, oligomer, and fibril samples containing Alexa Fluor 647-labeled αS at a ratio of 1:1,000 (labeled : unlabeled αS), were prepared as described in Section 2.4 αS Monomer, Oligomer and Fibril Preparation. Labeling of αS with Alexa Fluor 647 was conducted under a condition favoring conjugation at the αS N-terminus, as reported in our previous study in detail [12, 20]. Mixtures of labeled αS and PG65-MIMO were prepared as described in Section 2.11 Measurement of FlAsH Fluorescence. After allowing 1 hr for FlAsH binding, the αS and PG65-MIMO mixtures were loaded and ran on a native-PAGE gel. The control samples of αS alone and PG65 alone in the presence of FlAsH, EDT and TCEP were prepared and ran on a native-PAGE gel similarly. In-gel fluorescence imaging of the native-PAGE gel was captured on a GE Typhoon Trio phosphorimager at the Scientific Cores & Shared Resources at NYU Langone Medical Center.

2.13. Time-Course αS Aggregation

Lyophilized αS was dissolved in PBSA and subsequently filtered through a 0.45 μm syringe filter. The filtered αS solution at 70 μM was then incubated in a glass vial at 37 °C and under constant agitation by magnetic stirring at 100 rpm. Samples were aspirated at specific time points for characterization by ThT Fluorescence, FlAsH fluorescence, native-PAGE and TEM imaging.

3. RESULTS

3.1. Design of PG65-MIMO

We aimed to create a molecular probe for rapid and comprehensive reporting of αS aggregation states. To this end, we attempted to engineer PG65 (Fig. 1) to construct a PG65 variant that responds differently to distinct αS conformers. For the development of a molecular probe with multiple inputs and multiple outputs, we attempted to explore a linkage between conformational states and fluorescence signaling of PG65. The flanking residues of PG65’s tetra-cysteine motif were selected for this engineering effort, as these regions are critical in fluorescence signaling of PG65 through conformation-switching [13]. The importance of residues next to a tetra-cysteine motif in modulating FlAsH fluorescence signaling via binding to a target was also reported with a molecular probe, similar to PG65, but for detection of β-amyloid oligomers [21]. In the presented study herein, the four cysteines in the tetra-cysteine motif remained intact, as they are critical in FlAsH binding. The amino acids ‘PG’ in the tetra-cysteine motif, CCPGCC, which is widely believed as the minimal tetra-cysteine sequence for FlAsH binding [22], also remained unchanged because of high stability of the CCPGCC-FlAsH complex, when compared to other variations in the positions of PG [16, 17].

We then created a PG65 variant, PG65-MIMO (Fig. 1), by mutating the flanking region of the tetra-cysteine motif. Specifically, we replaced each of the three consecutive flanking residues, next to the N- and C-termini of CCPGCC, with HRW and KTF, respectively. HRW and KTF were chosen for the mutations, as the resulting HRWCCPGCCKTF sequence is an optimized sequence for FlAsH fluorescence [17]. While these mutations introduce significant changes in the flanking regions of PG65-MIMO, the number of amino acids remained same, when compared to PG65 (Fig. 1).

3.2. Characterization of PG65-MIMO

PG65-MIMO was found to be mostly unstructured and monomeric at 50 μM according to circular dichroism (CD) (Fig. 2A) and size exclusion chromatography (SEC) (Fig. 2B). PG65-MIMO exited the SEC column at the same elution volume as αS monomers, but with a higher A₂₈₀ (Fig. 2B) due to the presence of tryptophan in the new tetra-cysteine motif (Fig. 1). Additionally, PG65-MIMO’s molecular weight (MW) was similar to αS, as expected, with a theoretical MW of ~14.9 kDa and confirmed on SDS-PAGE (Fig. 2C). PG65-MIMO displayed a lower FlAsH fluorescence intensity when mixed with sodium dodecyl sulfate, due to changes in the secondary structure (Fig. 2D), indicating that PG65-MIMO’s FlAsH fluorescence is conformation-dependent.

Figure 2.

(A) CD spectra of PG65-MIMO in the 0 mM (red), 150 μM (black) and 1 mM (blue) TCEP. (B) SEC spectra of PG65-MIMO at 50 μM (blue) and αS monomers at 50 μM (M, black). (C) SDS-PAGE of αS (lane 1), PG65-MIMO (lane 2) and a Mark12 protein standard (lane 3). (D) FlAsH fluorescence and (inset) CD spectra of PG65-MIMO at 150 μM TCEP in the presence (blue) and absence (red) of 750 μM sodium dodecyl sulfate (SDS). (E) Tryptophan fluorescence (blue circles) and FlAsH fluorescence (red squares) of PG65-MIMO alone were measured at various TCEP concentrations. For reliable tryptophan fluorescence measurements, PG65-MIMO concentration was set at 0.5 μM. In (E), FlAsH fluorescence of PG65-MIMO (0.33 μM) with αS oligomers (2.5 μM) was measured at specific TCEP concentrations (50 μM, 75 μM, 100 μM and 150 μM; black triangles). Tryptophan fluorescence (λ_ex: 295 nm and λ_em: 350 nm) of PG65-MIMO with αS oligomers at a similar concentration ratio was not measured because of background fluorescence from αS oligomers with light in this wavelength range. In (E), results from FlAsH and tryptophan fluorescence experiments were normalized and adjusted for TCEP concentration per PG65-MIMO concentration. *: FlAsH fluorescence data obtained with 75 and 150 μM of TCEP. Error bars: 1 standard deviation of at least duplicates. In (D) and (E), the concentrations of PG65-MIMO and FlAsH were 0.33 and 1.5 μM, respectively, during FlAsH fluorescence measurements.

FlAsH fluorescence of PG65-MIMO was also measured at varying concentrations of TCEP (Fig. 2E). Interestingly, FlAsH fluorescence of PG65-MIMO appeared stronger with an increasing TCEP concentration, saturating at the TCEP/PG65-MIMO concentration ratio of ~200. To examine the origin of this TCEP dependency, we measured tryptophan fluorescence of PG65-MIMO at various concentrations of TCEP. As PG65-MIMO contains a single tryptophan within HRW, next to the tetra-cysteine motif (Fig. 1), its tryptophan fluorescence is strongly indicative of conformational states around the tetra-cysteine motif. The results show an overlapping trend for tryptophan and FlAsH fluorescence (Fig. 2E), suggesting that the conformation at the tetra-cysteine motif is TCEP concentration-dependent and directly related to PG65-MIMO’s FlAsH fluorescence. The conformational changes occurred locally near the tetra-cysteine motif, as no strong TCEP dependency was observed in the global secondary structure of PG65-MIMO (Fig. 2A). For subsequent experiments, we selected two TCEP concentrations (i.e., 75 and 150 μM for 0.33 μM of PG65-MIMO) corresponding to the TCEP/PG65-MIMO concentration ratio of ~230 and ~460 (shown with asterisks in Fig. 2E), which fall within the range of saturating FlAsH fluorescence. The concentration of PG65-MIMO was set at 0.33 μM, which is far below 50 μM used in SEC (Fig. 2B), during FlAsH fluorescence measurements in the presence and absence of αS (see Section 3.4) to ensure PG65-MIMO was monomeric during the measurements.

3.3. Characterization of preformed αS samples

Samples of the three major αS conformers, namely, monomers, oligomers and fibrils were prepared. The three conformers displayed unique structural and morphological properties, as reported elsewhere [8, 13, 20, 23–26]. The preformed αS monomers were structurally disordered and exhibited no significant fluorescence with Thioflavin T (ThT), a dye that detects αS fibrils over αS monomers and oligomers (Fig. 3A–B) [20]. Our αS monomers were not recognized by the A11 antibody (Fig. 3C), which specifically reacts to amyloid oligomeric conformations [27, 28]. The lack of any large aggregates in the preformed αS monomer samples was verified by transmission electron microscopy (TEM) (Fig. 3D). The preformed αS oligomers were rich in β-sheets, as characterized by a minimum at ~ 218 nm in CD (Fig. 3A) and displayed no strong ThT fluorescence (Fig. 3B). Our αS oligomers were recognizable by A11 (Fig. 3C) and showed globular and annular morphology with ~ 10 nm in diameter (Fig. 3D). The CD signals of preformed αS fibrils were weak due to low soluble αS concentration in these samples (Fig. 3A). Our αS fibril samples were ThT-positive, A11-negative and fibrillar in morphology (Fig. 3B–D). This result is largely consistent with αS fibril characterizations reported in the previous study [9], including those using the A11 antibody, which recognizes generic conformational epitopes that are present in oligomers of various amyloid proteins, but distinct from those displayed by amyloid fibrils [27, 28].

Figure 3.

(A) CD spectra, (B) ThT fluorescence, (C) A11 dot blot assay, and (D) representative TEM images of αS monomers (M), αS oligomers (O) and αS fibrils (F). In (D), the inset image of αS O sample focuses on the annular and globular morphology of αS aggregates. In (A), αS M: black, αS O: red, αS F: blue. In (B), ThT fluorescence of αS M, αS O, and αS F samples was measured at the same monomer equivalent concentration. Error bars: 1 standard deviation of triplicates.

3.4. FlAsH Fluorescence of PG65-MIMO with preformed αS

FlAsH fluorescence of PG65-MIMO was measured with preformed αS monomers, αS oligomers and αS fibrils at various αS and TCEP concentrations in the presence of excess FlAsH. Overall, PG65-MIMO responded uniquely to αS monomers, αS oligomers and αS fibril, and generated different relative FlAsH fluorescence change (RFC) profiles (Fig. 4).

Figure 4.

Relative FlAsH fluorescence changes (RFCs) of PG65-MIMO with (A) αS monomers (B) αS oligomers and (C) αS fibrils at 75 μM (red squares) and 150 μM (blue triangles) TCEP. (D) ΔRFC (= RFC at 150 μM TCEP – RFC at 75 μM TCEP) of PG65-MIMO with αS monomers (red circles), αS oligomers (blue diamonds) and αS fibrils (black squares). M: monomers, O: oligomers, F: fibrils. The dotted horizontal lines indicate either RFC or ΔRFC = 0. Error bars: 1 standard deviation of at least triplicate samples. In (D), errors of ΔRFC were determined using the propagation of error methods. One standard deviation of the average background ΔRFC = 0.1.

In the presence of αS monomers, PG65-MIMO generated positive RFC responses at the αS concentrations tested, with the RFC increase plateauing at ~1 μM of αS, independent of the TCEP concentration (Fig. 4A). Under our experimental set up, αS monomers at ≥ 0.25 μM were readily detectable by FlAsH fluorescence of PG65-MIMO. PG65-MIMO in the presence of αS oligomers displayed different FlAsH fluorescence, dependent on the concentration of TCEP in the assay buffer (Fig. 4B). At 75 μM of TCEP, there was a continuous decrease in PG65-MIMO’s RFC, which plateaued at ~3.5 μM of αS oligomers. At 150 μM of TCEP, RFCs of PG65-MIMO increased with an increasing concentration of αS oligomers in a similar manner to the response observed for αS monomers (Fig. 4B). We then extended monitoring of TCEP-dependent responses of PG65-MIMO (0.33 μM) to αS oligomers (2.5 μM) by measuring FlAsH fluorescence at two additional TCEP concentrations, 50 and 100 μM (Fig. 2E). Note that FlAsH fluorescence of PG65-MIMO alone was similar at 50 – 150 μM TCEP, corresponding to the concentration ratio of TCEP/PG65-MIMO at ~150 – ~ 450 (Fig. 2E). At the 50 and 75 μM of TCEP, FlAsH fluorescence intensity of PG65-MIMO was lowered when mixed with αS oligomers (Fig. 2E). At 100 μM TCEP (corresponding to the concentration ratio of TCEP/PG65-MIMO at ~300), FlAsH fluorescence of PG65-MIMO was similar whether mixed with αS oligomers or not (Fig. 2E). On the other hand, FlAsH fluorescence of samples containing PG65-MIMO and αS oligomers was stronger than PG65-MIMO alone at 150 μM TCEP (Fig. 2E). Thus, this result further confirms the TCEP-dependency of PG65-MIMO observed with αS oligomers. In contrast, no substantial RFC of PG65-MIMO was noticed with αS fibrils at ≤ 5 μM, regardless of the TCEP concentrations (Fig. 4C). In short, our results demonstrate that PG65-MIMO generates unique FlAsH fluorescence responses to different αS conformers depending on TCEP concentrations. The sensing behaviors of PG65-MIMO with αS monomers and oligomers seem unique, when compared to PG65: PG65 generated notable FlAsH fluorescence changes only with αS oligomers (Fig. S1), as reported previously [13], with no PG65-MIMO-like TCEP-dependency (Fig. S1). In all cases, detection of αS monomers and αS oligomers using FlAsH fluorescence of PG65-MIMO occurred within 2 hr, a notable improvement over an overnight-long, traditional ELISA-based detection.

As described above, PG65-MIMO’s RFC values were positive with both αS monomers and αS oligomers and ~0 with αS fibrils at 150 μM of TCEP, indicating that selective detection of soluble αS species (i.e., monomers and oligomers) is possible under this condition. Additionally, the observed αS conformer- and TCEP-dependency of PG65-MIMO may allow for αS oligomer-specific detection using the differences in RFC measured between at 150 and 75 μM of TCEP (ΔRFC) (Fig. 4D). ΔRFC of PG65-MIMO increased with an increasing concentration of αS oligomers and then level-offed at ~3.5 μM of αS oligomers, whereas ΔRFC was relatively small (~0) and constant with αS monomers and αS fibrils at the αS concentrations tested (Fig. 4D). Using ΔRFC of PG65-MIMO, αS oligomers at ≥ 0.5 μM were readily discernable from αS monomers and αS fibrils (Fig. 4D). The limit of detection (LOD) for αS oligomers by ΔRFC of PG65-MIMO was 0.5 μM, as judged by ≥ 3 standard deviations of the average background ΔRFC (Fig. 4D). This sensitivity represents a modest improvement over PG65, where the LOD was 1 μM [13]. In addition, αS monomer-selective detection is also possible under a specific condition (e.g., 75 μM of TCEP and αS ≤ 0.5 μM, where PG65-MIMO generated non-negligible signals with only αS monomers). Taken together, our results show that detection of αS soluble fractions over αS insoluble fractions (i.e., fibrils) as well as further distinction between αS monomers and αS oligomers is possible with FlAsH fluorescence and TCEP-dependence of PG65-MIMO.

3.5. Time-Course Aggregation of αS probed by PG65-MIMO

Encouraged by the observed αS conformer-specific and TCEP-dependent FlAsH fluorescence of PG65-MIMO, we tested if one can use the dual TCEP FlAsH fluorescence assay to evaluate aggregation states in αS samples during incubation over time (Fig. 5A–B). For comparison, aggregation of the αS samples was also examined using ThT, which displayed fluorescence with αS fibrils > 100-times more strongly than with αS monomers and αS oligomers (Fig. 3B). For this examination, 70 μM of freshly prepared αS solution (containing none of PG65-MIMO, FlAsH and ThT), which lacked any large aggregates (Fig. 5C), was incubated at 37 °C with constant stirring. Aliquots were withdrawn during the incubation and mixed with ThT. As reported elsewhere [13, 25, 29], ThT fluorescence on the incubated samples followed a typical sigmoidal change, with little to no fluorescence until Day 1, followed by a rapid increase and reaching a maximum on Day 3 of incubation (Fig. 5A). The presence of αS fibrils in samples on Day 5 was verified by TEM (Fig. 5C). Additional aliquots of αS samples were taken during the incubation and diluted into the assay buffer containing PG65-MIMO (0.33 μM), FlAsH (1.5 μM) and TCEP (75 or 150 μM). The RFC of PG65-MIMO at 150 μM of TCEP was ~1.2 with aliquots of the αS samples (2.5 μM) on Day 0 (Fig. 5B). This RFC value was ~20 % higher than that obtained with 2.5 μM of αS monomers or αS oligomers (Fig. 4A–B). While the origin of the higher RFC is currently elusive, the RFC data unequivocally indicate that αS samples freshly prepared on Day 0 contained predominantly αS soluble species (i.e., monomers and oligomers). ΔRFC of PG65-MIMO, an indicator of αS oligomer concentration, was low (~0.05) with αS samples on Day 0 (Fig. 5A). Then, ΔRFC of PG65-MIMO increased with αS samples taken after longer incubation until Day 2 (where ΔRFC = ~0.45 corresponding to ~0.4 μM of αS oligomers in αS aliquots (2.5 μM), see Figs. 4D and 5A), with the value remaining similar thereafter (Fig. 5A). Consistent with the time-course ΔRFC change of PG65-MIMO, αS oligomerization was evident during the incubation, when judged by native-PAGE with in-gel fluorescence (Fig. 5D): the streaking bands in the upper portion of the gel indicative of αS oligomers [20] (also see Section 3.6), located above the primary αS monomeric bands [20] (also see Section 3.6), became stronger over time and the oligomeric band intensities leveled off on Day 2 (Fig. 5D). The presence of oligomeric αS aggregates on Day 2 was also confirmed by TEM (Fig. 5C). The RFCs of PG65-MIMO at 150 μM TCEP, a measure of αS soluble species concentration, decreased with αS samples taken during incubation over time (Fig. 5B). This result is in good agreement with the weakened intensities of bands for αS species that entered the native-PAGE gel (Fig. 5D). Collectively, consistent with the previous findings as well as the current ThT, TEM and native-PAGE data, our FlAsH fluorescence data indicate that (1) freshly prepared αS samples were mostly monomeric, (2) additional oligomerization occurred to a moderate extent (~16%) upon prolonged incubation of fresh αS samples [6–8, 12, 18, 23], and (3) the soluble αS species was converted to insoluble fibrils during the incubation [12, 13, 18].

Figure 5.

Time-course aggregation of αS monitored by (A) ThT fluorescence (blue circles) and ΔRFC (= RFC at 150 μM TCEP – RFC at 75 μM TCEP, red diamonds) of PG65-MIMO, (B) FlAsH fluorescence of PG65-MIMO at 75 (red squares) and 150 μM (blue triangles) TCEP concentrations, (C) TEM and (D) native-PAGE with in-gel fluorescence. Freshly prepared αS was incubated at 70 μM and 37 °C, in a glass vial with constant magnetic stirring. Aliquots of αS were withdrawn at designated time points to analyze the aggregation state of αS using ThT and FlAsH fluorescence, native-PAGE and TEM. TEM was used for qualitative analyses of αS morphology. In (A) and (B), error bars: 1 standard deviation of duplicate. Errors of ΔRFC were determined using the propagation of error methods. In (D), αS samples contained Alexa Fluor 647-labeled αS for sensitive monitoring of αS oligomerization. The image in the right provides brighter upper portions of the native-PAGE gel – indicated by the bracket – obtained by renormalizing the brightness from the lower to the upper bands, which often contained fewer percentages of fluorophores.

3.6. Binding between PG65-MIMO and αS

We then examined if the observed FlAsH responses of PG65-MIMO stemmed from its binding to αS. The binding between PG65-MIMO and αS was examined by native-PAGE. For specific monitoring of αS on a native-PAGE gel, αS was labeled with Alexa Fluor-647. Samples of preformed αS monomers, oligomers and fibrils containing the labeled αS (at a concentration ratio of labeled/unlabeled αS = 1:1,000) were prepared accordingly. Our previous study demonstrated that the labeled αS conformers prepared in this manner exhibit aggregation properties and electrophoretic mobility similar to those exclusively unlabeled [13, 20]. PG65-MIMO was selectively monitored on a native-PAGE gel using FlAsH fluorescence, which resulted from its binding to FlAsH. To mimic FlAsH fluorescence assay conditions, samples of PG65-MIMO (0.33 μM) and/or αS (2.5 μM) were mixed with FlAsH (1.5 μM) and TCEP (75 or 150 μM) prior to running on the gel. Note that we did not attempt to determine molecular weights of αS and PG65-MIMO on a native-PAGE gel using molecular weight standards, as protein migration on a native-PAGE gel depends not only on molecular weight, but also charge and conformation of the protein.

Relative positions on a native-PAGE gel of our labeled αS monomers, oligomers, and fibrils were generally consistent with previous reports [9, 20, 30] (Fig. 6A–B). Band patterns and their fluorescence intensities were similar at 75 and 150 μM TCEP concentrations for each of these αS control samples. αS monomer bands are located at the lower portion of the native-PAGE gel. On the other hand, streaking bands with strong intensities at the upper and lower portions of the native-PAGE gel were found with αS oligomer samples. In contrast, αS fibrils were trapped in the well, not entering the gel. Though PG65-MIMO is monomeric (Fig. 2B), FlAsH-bound PG65-MIMO appeared as streaking bands on the native PAGE gel unlike αS monomers (Fig. 6A–B, lane 7 vs. lane 1). The implication is that FlAsH binding affects electrophoretic mobility of PG65-MIMO. The band patterns and their fluorescence intensities of PG65-MIMO alone were similar at the two TCEP concentrations, consistent with the FlAsH fluorescence trend of PG65-MIMO controls in this TCEP concentration range (Fig. 2E). We used the native-PAGE gel for qualitative rather than quantitative analyses, as accurate quantification of fluorescence intensities, in particular, of streaking bands are difficult. Visualization using green fluorescence showed streaking bands at the upper portion of the gel in the αS oligomer samples (enclosed by white rectangles in Fig. 6A–B, lane 2), indicating background FlAsH fluorescence of these samples due to non-specific binding of FlAsH. No such background FlAsH fluorescence was observed with other αS control samples (Fig. 6A–B, lanes 1 and 3). It should be noted that background FlAsH fluorescence of αS (e.g., ~20 % of FlAsH fluorescence of PG65-MIMO (0.33 μM) from 2.5 μM αS oligomers) was taken into account and subtracted during calculation of RFCs in the aforementioned FlAsH fluorescence assays (Fig. 4; also see Section 2.11).

Figure 6.

In-gel fluorescent imaging of native-PAGE gels on Alexa Fluor 647-labeled αS (red), FlAsH-bound PG65-MIMO (Green) and their mixtures at (A) 75 and (B) 150 μM TCEP. Top: overlay. Middle: green filtered channel. Bottom: red filtered channel. αS M: αS monomers. αS O: αS oligomers. αS F: αS fibrils. Some bands are enclosed by white rectangles and circles (see Section 3.6).

Samples containing both PG65-MIMO and labeled αS in the presence of FlAsH and TCEP were also resolved on a native-PAGE gel and monitored using fluorescence (Fig. 6A–B, lanes 4–6). When PG65-MIMO and αS monomers were mixed, new overlapping bands were observed on the gel, demonstrating binding between the two entities (enclosed by white circles in Fig. 6A–B, lane 4). These overlapping bands showed strong green fluorescence at the two TCEP concentrations, responsible for the positive RFCs of PG65-MIMO with αS monomers independent of the TCEP concentrations (Fig. 4A). The addition of PG65-MIMO (0.33 μM) to αS monomers (2.5 μM) did not seem to cause multimerization of these αS species, as judged by the lack of any new streaking αS bands (red) formed on the native-PAGE gels (Fig. 6A–B). No evidence of αS multimerization was observed either at higher PG65-MIMO to αS monomer concentrations ratios (Fig. S2). Instead, αS monomer bands were shifted down in the presence of equimolar or excess PG65-MIMO on the native-PAGE gel (Fig. S2). The band shift is likely due to a change in structural flexibility of αS monomers upon binding to PG65-MIMO, from a highly disordered conformation (corresponding to a hydrodynamic size of ~60 kDa globular proteins [5]) to a relatively compact state. New overlapping bands were also observed from samples of PG65-MIMO and αS oligomers at the lower portion of the gel (enclosed by white circles in Fig. 6A–B, lane 5). Interestingly, green fluorescence intensities of these overlapping bands were stronger at 150 μM than 75 μM TCEP, accounting for the observed PG65-MIMO’s TCEP dependency with αS oligomers (Fig. 4B). This finding also suggests that these overlapping bands did not result from binding between PG65-MIMO and αS monomers (possibly present as a minor fraction in αS oligomer samples). This is because green fluorescence intensities of the overlapping bands significantly differed at the two TCEP concentrations, unlike those observed from the mixture samples of PG65-MIMO and αS monomers (Fig. 6A–B). Moreover, green fluorescence bands at the overlapping regions from the mixture samples of PG65-MIMO and αS oligomers were located slightly above those from the mixture samples of PG65-MIMO and αS monomers at both TCEP concentrations (see white circles in lanes 4 and 5 of Fig. 6A–B). Though the exact identity of the overlapping bands between PG65-MIMO and αS oligomers has yet to be determined, these bands might indicate binding of PG65-MIMO to a subset of the αS oligomers. For this mixture, the green streaking bands were overlapped with the red streaking bands in the upper portion of the gel (see white rectangles in lane 5 of Fig.6 A–B). Green fluorescence intensities of these bands were similar to those of αS oligomers only (enclosed by white rectangles in lane 2 of Fig. 6A–B) because of either excess free αS oligomers remaining unbound to PG65-MIMO or no significant FlAsH fluorescence change of PG65-MIMO upon the binding event indicated by the upper overlapping bands (enclosed by white rectangles in lane 5 of Fig. 6A–B). In contrast to αS monomers and αS oligomers, no notable binding between PG65-MIMO and αS fibrils was observed at the two TCEP concentrations, as judged by the lack of significant electrophoretic pattern change of PG65-MIMO (Fig. 6A–B, lane 6 vs. lane 7). This result is consistent with negligible RFCs of PG65-MIMO with αS fibrils regardless of the TCEP concentrations (Fig. 4C).

4. DISCUSSION

Our data suggest that PG65-MIMO can produce multiple outputs (i.e., TCEP-dependent FlAsH fluorescence signals) in response to multiple inputs (i.e., αS monomers and αS oligomers) via a conformation-driven mechanism. An intrinsically disordered protein, such as αS, often interacts with multiple partners, playing an important role in signaling and regulatory processes [31, 32]. PG65-MIMO shares the same amino acid sequence with αS except for the replacement of αS 62–71 (QVTNVGGAVV) with HRWCCPGCCKTF (Fig. 1). PG65-MIMO is structurally disordered and monomeric (Fig. 2A–B). Thus, PG65-MIMO is likely to adopt significant conformational flexibility as intrinsically disordered αS, pointing that αS variants may serve as useful structural prototypes, which can readily evolve into molecular probes for handling multiple inputs and multiple outputs. The conformation-sensitive FlAsH fluorescence also plays a critical role in the connection between inputs and outputs. For this connection, a tetra-cysteine motif would need to be properly placed at a location where non-negligible conformational changes occur upon receiving inputs (e.g., binding to targets). αS 62–71 undergoes significant structural changes during self-assembly of αS monomers into oligomers, and into fibrils [9, 33, 34]. Thus, the HRWCCPGCCKTF sequence of PG65-MIMO would experience similar conformational changes upon binding to αS. On the other hand, FlAsH fluorescence of αS variants with the tetra-cysteine motif located at its N-terminus [35] and C-terminus [36, 37] appeared independent of its self-assembly, suggesting that conformational states at the N-terminal and C-terminal regions remain similar during aggregation. We note that engineering of the residues that flank the tetra-cysteine motif can create novel sensing characteristics, such as the TCEP-dependency observed with PG65-MIMO (containing HRWCCPGCCKTF), but not with PG65 (containing QVTCCPGCCAVV). αS detection was possible within 2 hr using FlAsH fluorescence of PG65-MIMO, in contrast to an overnight-long, traditional ELISA-based method. It should however be noted that rapid αS oligomer detection can also be achieved by adopting a microfluidic platform for an immuno-probe [38] or using chemical probes [39] or a combination thereof, such as phenylene ethynylene derivatives and ThT [40]. Nevertheless, the sensing characteristics of PG65-MIMO for producing different signal outputs depending on αS conformers makes it serve as a valuable probe that complements the current techniques to detect αS oligomers. The LOD for αS oligomer detection by ΔRFC of PG65-MIMO was 0.5 μM, an αS concentration close to physiological relevance (~1 μM) [41, 42]. The sensitivity also represents a slight improvement over PG65, where the LOD was 1 μM [13]. While the improvement is modest, more extensive sequence exploration of the flanking residues of PG65-MIMO, followed by FlAsH fluorescence-based evaluation of combinatorial probe libraries, has high potential for sensitivity improvement for the following reasons: (1) αS residues corresponding to the flanking residues of PG65-MIMO are critical in αS self-assembly [43, 44], making sequence optimization of PG65-MIMO at this location possibly increase its affinity for αS oligomers and (2) FlAsH fluorescence itself directly depends on the flanking residues of the tetra-cysteine motif [16, 17]. Collectively, re-optimization of the flanking residues via random mutagenesis and fluorescence-based screening would lead to (1) the evolution of PG65-MIMO into other molecular probes with different inputs and outputs and (2) the identification of sequence characteristics optimal for αS oligomer binding-driven, sensitive FlAsH fluorescence signaling.

The primary application of PG65-MIMO would be an in vitro assay for evaluation of αS aggregation states in samples, for example, to rapidly evaluate impact of small molecules on αS conformers. Applications of PG65-MIMO for biological samples under complex conditions (e.g., with cell lysate, plasma proteins or other biopolymers) require careful examination of non-specific interactions of PG65-MIMO and/or FlAsH with the cellular components. In particular, non-specific binding of FlAsH to other amyloid oligomers or endogenous proteins abundant in cysteines might occur, complicating interpretation of the detection results. Thus, to minimize the background FlAsH fluorescence signals, concentrations of EDT and TCEP need to be re-optimized. Calibration of FlAsH fluorescence responses of PG65-MIMO with biologically-derived αS samples might also be necessary to quantify αS conformers in biological samples.

As judged by tryptophan and FlAsH fluorescence, the local conformational states of PG65-MIMO at or near its tetra-cysteine motif can be changed by the addition of TCEP (Fig. 2E). Among different chemical moieties of TCEP, its three carboxyl groups might be responsible for the conformational changes of PG65-MIMO. The carboxyl groups can interact with the positively charged N-terminus (M1-K60) of PG65-MIMO (+4.1 at pH 7). Importantly, solvent exposure of the same N-terminus (M1-K60) of αS, which interacts with the C-terminus (K96-A140) of αS, determines conformations of monomeric αS [45]. Thus, TCEP may affect the conformational states of PG65-MIMO by interacting with its N-terminus. Our results show that PG65-MIMO binds to αS monomers and αS oligomers and, upon binding, exhibits FlAsH fluorescence changes. The observed electrophoretic patterns of PG65-MIMO on a native-PAGE gel in the presence of αS monomers and αS oligomers, when compared to their absence (Fig. 6A–B), suggest the high likelihood of conformational changes in PG65 upon the binding, which were translated to RFCs (Fig. 4A–B). The RFCs of PG65-MIMO were similar with αS monomers at the two TCEP concentrations (75 and 150 μM), suggesting that the binding-induced conformational changes occur to a similar extent regardless of the TCEP concentrations. Although overall TCEP concentration-dependent, FlAsH fluorescence of PG65-MIMO alone (0.33 μM) was saturated at 50–150 μM of TCEP (Fig. 2E). However, αS oligomer binding to PG65-MIMO seemed to push this TCEP dependency of FlAsH fluorescence to a higher TCEP concentration (compare red squares and black triangles in Fig. 2E), resulting in negative and positive RFCs of PG65-MIMO at 75 and 150 μM of TCEP, respectively (Fig. 4B). While the exact origin of the PG65-MIMO’s TCEP dependency with αS oligomers is not fully understood, binding to oligomeric rather than monomeric αS might limit access of TCEP to the N-terminus of PG65-MIMO, increasing this probe’s resistance to conformational changes by TCEP. αS oligomer concentrations for RFC plateaus were different at the two TCEP concentration (Fig. 4B). This might be because different regions on αS oligomers were utilized for binding of PG65-MIMO, depending on TCEP concentrations.

Our observation that ΔRFC increased before a significant ThT fluorescence increase (Fig. 5A) indicates that αS oligomerization precedes αS fibrillization. In addition, our preformed αS oligomers, which were detectable by PG65-MIMO, seeded αS fibrillization [13, 20]. These observations are in good agreement with the view that αS oligomers detectable by PG65-MIMO may be intermediates on the αS fibrillization pathway. As judged by ΔRFC, depletion of the on-pathway αS oligomers was slow (Fig. 5A). This might be due to relatively high kinetic stability of the αS oligomers, as reported elsewhere under similar experimental conditions [18, 46]. Note that not every type of αS oligomers found in vivo and in vitro are necessarily detectable by PG65-MIMO. An additional biological examination in combination with fluorescent assays using PG65-MIMO will be required to better understand the biological nature of PG65-MIMOd-etectable αS oligomers.

Multiple lines of studies suggests that αS monomers can directly bind to αS fibrils [47, 48]. Interestingly, we observed no notable binding of PG65-MIMO to αS fibrils. Apparently, the lack of such binding is due to inclusion of the HRWCCPGCCKTF sequence in PG65-MIMO in lieu of αS 62–71. The impact of the mutation on αS fibril binding might be related to the unique parallel, in-register, intermolecular β-sheet association of αS monomeric units within αS fibrils. αS 62–71 is one of the four regions with high parallel, in-register β-sheet propensities and mutations in this region may interfere with self-assembly into fibrils [12]. Thus, PG65-MIMO may have limited ability to associate with αS fibrils. No such parallel-in-register β-sheets are formed around αS 62–71 during αS assembly into dimers and oligomers [9, 49–52], allowing binding of PG65-MIMO to αS monomers and oligomers.

5. CONCLUSIONS

In summary, we describe the creation of a protein probe for rapid and comprehensive evaluation of αS aggregation states by responding differently to distinct αS conformers. We anticipate that our probe will be useful in understanding a link between αS aggregation profiles and PD pathology and developing and identifying aggregation inhibitors of therapeutic relevance.

HIGHLIGHTS.

• PG65-MIMO is an engineered molecular probe derived from α-synuclein (αS).

• Rapid and comprehensive profiling of αS aggregation is possible with PG65-MIMO.

• PG65-MIMO generates distinct fluorescence responses to the three major αS conformers.

• Fluorescence signaling of PG65-MIMO in response to αS oligomers is TCEP-dependent.

• PG65-MIMO’s conformational states determine its fluorescence signaling.

ABBREVIATIONS

αS

α-synuclein

PD

Parkinson’s disease

NAC

non-amyloid-β component

ELISA

enzyme-linked immunosorbent assay

TCEP

tris(2-carboxyethyl)phosphine

FPLC

Fast protein liquid chromatography

PBSA

phosphate buffered saline with azide

CD

circular dichroism

ThT

Thioflavin T

SEC

size exclusion chromatography

TEM

transmission electron microscopy

SDS

sodium dodecyl sulfate

PAGE

polyacrylamide gel electrophoresis

RFC

relative FlAsH fluorescence change