Comparative Analysis of the Relative Fragmentation Stabilities of Polymorphic Alpha-Synuclein Amyloid Fibrils

Abstract

The division of amyloid fibril particles through fragmentation is implicated in the progression of human neurodegenerative disorders such as Parkinson's disease. Fragmentation of amyloid fibrils plays a crucial role in the propagation of the amyloid state encoded in their three-dimensional structures and may have an important role in the spreading of potentially pathological properties and phenotypes in amyloid-associated diseases. However, despite the mechanistic importance of fibril fragmentation, the relative stabilities of different types or different polymorphs of amyloid fibrils toward fragmentation remain to be quantified. We have previously developed an approach to compare the relative stabilities of different types of amyloid fibrils toward fragmentation. In this study, we show that controlled sonication, a widely used method of mechanical perturbation for amyloid seed generation, can be used as a form of mechanical perturbation for rapid comparative assessment of the relative fragmentation stabilities of different amyloid fibril structures. This approach is applied to assess the relative fragmentation stabilities of amyloid formed in vitro from wild type (WT) α-synuclein and two familial mutant variants of α-synuclein (A30P and A53T) that generate morphologically different fibril structures. Our results demonstrate that the fibril fragmentation stabilities of these different α-synuclein fibril polymorphs are all highly length dependent but distinct, with both A30P and A53T α-synuclein fibrils displaying increased resistance towards sonication-induced fibril fragmentation compared with WT α-synuclein fibrils. These conclusions show that fragmentation stabilities of different amyloid fibril polymorph structures can be diverse and suggest that the approach we report here will be useful in comparing the relative stabilities of amyloid fibril types or fibril polymorphs toward fragmentation under different biological conditions.

Keywords: amyloid; atomic force microscopy; fibril division; fibril fragmentation; image analysis; sonication; stability.

Figure 1

WT, A30P, and A53T α-synuclein amyloid fibrils formed in vitro display different fibril morphologies. AFM images of amyloid fibrils samples formed from WT (a,b), A30P (d,e) and A53T (g,h) α-synuclein. All three variants of α-synuclein amyloid fibrils were formed at 50 µM monomer equivalent concentration by gentle shaking at 180 rpm, 37 °C, for at least 14 days. AFM images representing 10 × 10 µm (a,d,g) or 4 × 4 µm (b,e,h) surface areas are shown with the scale bars indicating 2 µm. The white arrowheads in panels (b,e) and (h) indicate individual fibrils that are further magnified and shown in panels (c,f,i), respectively. In panels (c,f,i), the digitally straightened fibril images are shown together with 3D-reconstructed surface envelope models, their central-line height profiles (blue lines) and the population average height (pale blue lines) to demonstrate differences in fibril morphology.

Figure 2

AFM imaging of WT, A30P and A53T α-synuclein amyloid fibrils undergoing fragmentation by controlled sonication. WT (a), A30P (b) and A53T (c) α-synuclein amyloid fibril samples formed from 50 µM monomer equivalent protein concentration were sonicated for up to 640 s, with the duration of sonication indicated for each time-point sample. AFM images representing 10 × 10 µm surface areas are shown with insets of 5× magnified images. The scale bars indicate 2 µm in all images.

Figure 3

Fibril length and height distributions extracted from AFM images of the WT, A30P and A53T α-synuclein amyloid fibrils undergoing fragmentation by controlled sonication. Histograms representing normalised fibril length distributions (a) and height distributions (c) are shown using the same length and height scales, respectively, for comparison. Cumulative distribution functions of the same length distributions (b) and height distributions (d) are also shown to facilitate visualisation of the changes in the length distributions and the consistency of the height distributions for the duration of the controlled sonication. Statistics of the quantitative image analysis are shown in Supplementary Table S1.

Figure 4

Comparative analysis of the relative stabilities of WT, A30P and A53T α-synuclein amyloid fibrils toward sonication-induced fragmentation. (a) The decay of experimentally observed mean lengths (+) as a function of sonication time duration is shown in a log–log plot together with the best-fit pure fragmentation model lines [26]. The thicker portion of the lines denotes the time ranges where the best-fit asymptotic lines describing characteristic length decay have likely been reached in the imaging experiments (the time ranges where the best-fit pure fragmentation model is likely reliable in describing the experimentally observed length decay, t > t_s in Equation (1)). (b) The fragmentation rate constants, B(x), obtained from the best-fit pure fragmentation models are shown in a log–log plot as functions of fibril lengths, x. The thicker portion of the lines denotes the range of fibril lengths observed experimentally on the AFM images.