Diverse Metastable Structures Formed by Small Oligomers of α-Synuclein Probed by Force Spectroscopy

Oligomeric aggregates are widely suspected as toxic agents in diseases caused by protein aggregation, yet they remain poorly characterized, partly because they are challenging to isolate from a heterogeneous mixture of species. We developed an assay for characterizing structure, stability, and kinetics of individual oligomers at high resolution and sensitivity using single-molecule force spectroscopy, and applied it to observe the formation of transient structured aggregates within single oligomers of α-synuclein, an intrinsically-disordered protein linked to Parkinson’s disease. Measurements of the molecular extension as the proteins unfolded under tension in optical tweezers revealed that even small oligomers could form numerous metastable structures, with a surprisingly broad range of sizes. Comparing the structures formed in monomers, dimers and tetramers, we found that the average mechanical stability increased with oligomer size. Most structures formed within a minute, with size-dependent rates. These results provide a new window onto the complex α-synuclein aggregation landscape, characterizing the microscopic structural heterogeneity and kinetics of different pathways.

. SDS-PAGE of protein constructs  Table   Table S1. Summary of potential unfolding distances estimated from structural models in literature

Aggregation of tandem-repeat constructs
The ability of the tandem-repeat α-synuclein constructs to aggregate into amyloid fibers was tested by measuring ThT fluorescence similar to standard methods [73]. Briefly, 50 μM tetrameric α-synuclein and 40 μM ThT dye were incubated at 37°C and shaken linearly at 20 and 30 Hz for 10 days. The ThT fluorescence, excited at 430 nm, was measured at 485 nm in a microplate reader (Gemini EM, Molecular Devices). As seen in Figure S2, the ThT fluorescence displayed the typical sigmoidal rise that occurs during amyloid formation after a lag time of approximately two days.

CD spectra of monomer, dimer, and tetramer
To assess the average structural content of the tandem-repeat oligomers, we measured the CD spectra of the three protein constructs (Jasco J-810 CD/ORD spectrometer) at a concentration of 10 μM in 10 mM phosphate buffer pH 7.0, using a 1-mm path length and subtracting the baseline spectrum of the buffer. All three spectra, shown in Figure S4, were characteristic of a largely disordered protein and similar to results found previously for αsynuclein [73].

Interpretation of unfolding distances
It is difficult to assign specific structures to particular ΔL c values without additional information, especially given the large number of transitions observed here. However, we have catalogued some of the various different structures that have been observed or proposed for α-synuclein and listed the associated contour length changes expected for unfolding (Table S1). Contour length estimates are based on the number of amino acids involved in the structures and the distance between the points at which force is applied to the protein (as estimated from the structural models).
Monomers of α-synuclein have been observed to form α-helical structures at the N-terminus under certain conditions [74][75][76]: an extended α-helix of ~ 90 aa, or two broken α-helices -one ~ 35 aa long and the other ~ 48 aa, folded into a helix-turn-helix "hairpin". Since helices have a length of 0.15 nm/aa [76], unfolding these α-helices would give rise to ΔL c ~ 20 nm for the extended helix, ~ 7 and 10 nm for the short helices individually if not folded into a hairpin, and ~ 30 nm for the full hairpin. Suggestively, transitions with ΔL c ~ 10, 20, and 30 nm are all seen for each of the constructs (monomer, dimer, and tetramer). Interestingly, the unfolding transitions with ΔL c ~ 30 nm seen in FECs of the monomer match the results of a previous AFM study (ΔL c ~ 28 nm), which the authors suggested arose from β-sheet structures in the N-terminus [77].
The structure of the amyloid fibril formed by αsynuclein[78] involves each monomer forming a 5strand β-sandwich, with individual sandwiches then aligned side by side. Many different ΔL c values could be expected from unfolding different components of such β-sandwiches, whether the sandwiches are in isolation or stacked in multimers. For example, unfolding of different numbers of β-strands in one or more sandwiches would lead to ΔL c ~ 8-10 nm for 2-3 strands or ~ 16-19 nm for 4-5 strands from one or two sandwiches, separation of neighboring sandwiches would produce ΔL c ~ 33-34 nm, and unfolding of one complete monomer from a stack of sandwiches would produce ΔL c ~ 50 nm. Some of these unfolding transitions are illustrated in Figure  S5. Several different transitions could in principle occur in combination, depending on the size of the construct (dimer or tetramer), giving rise to the large S3 number of ΔL c values listed in Table S1. Intriguingly, many of these values coincide with transitions observed in the FECs, suggesting that amyloid-like structural motifs may form even in the smallest oligomers.
A tetrameric native structure for α-synuclein was recently reported [79,80]. The structural model involves monomers folded into helix-hairpins similar to the micelle-bound structure, stacked in parallel. As for the β-sandwich structure, multiple ΔL c values could be expected from unfolding different combinations of structural components. Many of these values would be degenerate with the ΔL c values expected from other structures (e.g. 50 nm for removing a complete monomer, ~ 30 nm for unfolding a single helix-hairpin, ~ 18-19 nm for separating two neighboring helix-hairpins). ΔL c for complete unfolding of a tetramer, trimer, or dimer are quite distinct, however, being ~ 185 nm, ~ 135 nm, and ~ 85 nm, respectively.
The ability to assign specific structures to the unfolding transitions would provide deeper insight into the conformational changes of α-synuclein. One approach that may prove fruitful for such identifications is comparison to molecular dynamics simulations. Recent work showed how such simulations can be used to help interpret singlemolecule fluorescence studies of intrinsicallydisordered proteins [81]. However, simulations of oligomers have not yet been done, and may present challenges owing to the large size of the system being simulated.
As a final note, α-synuclein has been shown to bind non-specifically to dsDNA [82,83]. Hence it is possible that the sawtooth patterns in the FECs reflect dissociation of α-synuclein bound to the DNA handles rather than cooperative unfolding transitions. If this were the case, however, we would expect to observe a smooth distribution of ΔL c values, since non-specific interactions will not give rise repeatedly to the same discrete values. The fact that the actual distribution is highly peaked indicates that the transitions do indeed arise from protein structures, rather than protein-handle interactions.          Table   Table S1. Summary of potential unfolding distances estimated from structural models in literature. a Contour length changes expected from unfolding different structural components of the stacked, 5-stranded βsandwich structure in amyloid fibrils of α-synuclein, and the proposed α-helical tetramer of native α-synuclein. For the β-sandwich, ΔL c values include unfolding of various β-strands, unstacking of β-sandwiches, and unfolding of complete monomers. b For the helical tetramer, ΔL c values include unfolding individual helices, unfolding hairpins consisting of the N-and C-terminal helices, unstacking of neighboring helix-hairpins, and unfolding of complete monomers. Different permutations of these transitions are also possible.