The role of RNA in the nanoscale organization of α-synuclein phase separation

Abstract

The cellular accumulation of α-synuclein (aS) aggregates is a hallmark of several neurodegenerative diseases. Recent studies suggest that the aberrant transition of monomeric aS into solid-like aggregates may occur through an intermediate liquid-like state, where the protein partitions between dense and dilute phases. Although aS is not typically recognized as an RNA-binding protein, it can bind RNA under aggregation conditions, but its impact on aS liquid-like phases remains unexplored. Employing a combination of fluorescence spectroscopy techniques, we investigated aS mobility in both phases in the presence of RNA. Our analysis revealed the formation of nanoclusters involved in initiating phase separation and uncovered heterogeneity within the dense phase, discovering that aS molecules exist in two distinct mobility states. Additionally, we demonstrated that RNA induces morphological changes and promotes the liquid-to-solid transition of aS dense phase. These findings underscore the active role of RNA in modulating aS phase transitions.

Graphical Abstract

Figure 1.

Schematic representation of the experimental and analysis session. The phase separation assay is a mixture of aS, aS-atto488, and, in the case of the RNA experiment, total yeast RNA. The sample is placed into a closed chamber and under a confocal laser scanning microscope equipped with a 5 × 5 single-photon array detector. Fluorescence over time is acquired in different sample positions. The same dataset of fluorescence over time can be used to perform fluorescence lifetime quantification or FCS-based methods (FCS and spot-variation FCS).

Figure 2.

aS forms nano- and micro-complexes in a phase-separated system. (A) Confocal image (top) of 700 nM monomeric aS-atto488 in PB. Phase contrast microscopy image (middle) and confocal (bottom) image of aS in LLPS buffer. In the presence of 20% PEG, aS spontaneously forms a dense phase surrounded by a dilute phase. Scale bars: 5 μm. (B) Representative autocorrelation functions in the dilute and the dense phases of aS, as indicated by the arrows in panel (A). The lines represent the fit of the autocorrelation curves with a two-component fitting model. Diffusion times for the fast (τ_D1) and slow (τ_D2) diffusing components are shown. (C) Histogram of diffusion times τ_D1 measured for the fast diffusing component in the dense and dilute phases. The solid lines represent the upper envelope of the histograms. The dashed lines represent the median of both distributions; the vertical solid line indicates the median of the τ_D for monomeric aS in the LLPS buffer. n = 50 (dilute phase), 53 (dense phase). (D and E) svFCS analysis of the fast (D) and slow (E) diffusing components for both dilute and dense phases. Diffusion times are plotted against the observation areas ω₀² to retrieve the diffusion law τ_D(ω₀²) (dashed line). Data points and error bars are, respectively, mean and standard deviations of multiple measurements (n = 4 independent experiments). (F) Percentage of the fast diffusing component in the dilute and the dense phases. n = 40 (dilute phase), 48 (dense phase). Mann–Whitney statistical test (*: P-value < 0.05). (G) Box-plot of diffusion times τ_D2 measured for the slow diffusing component in the dilute and the dense phase. n = 37 (dilute phase), 42 (dense phase). Mann–Whitney statistical test (*: P-value < 0.05). The horizontal line in each box represents the median, the edges are the 25th and 75th percentile, and the vertical line extends to the minimum and maximum data points.

Figure 3.

Effect of total yeast RNA on aS phase-separated system. (A) On the left, confocal images of 5, 50, and 500 ng/μl RNA in LLPS buffer. On the right, confocal images of aS, aS-atto488 and RNA at different concentrations of RNA. A merge of both channels is shown on the right. Scale bars: 5 μm. (B and C) svFCS analysis of the fast (B) and slow (C) diffusing components of aS in the presence of RNA within the dilute phase. Diffusion times τ_D are plotted against the observation areas ω₀². Data points and error bars are, respectively, mean and standard deviations of multiple measurements (n = 4 independent experiments). (D) Histogram of diffusion times τ_D for the fast diffusing component of aS in the dilute phase under LLPS condition alone and in the presence of RNA (500 ng/μl). The solid lines on top of the histograms represent the upper envelope of the histogram. The dashed lines represent the medians of both distributions (1.88 ms for RNA); the vertical solid line indicates the median of the τ_D measured for the monomeric aS. n = 50 (aS), 57 (aS + RNA). (E) Percentage of the fast diffusing component in the dilute phase of aS in the presence of RNA. From left to right, n = 40, 30, 31, 77. Kruskal–Wallis statistical test followed by Dunn’s test (****P-value < 0.00005). (F) Box plots of the hydrodynamic radius (r_H) in the dilute phase of aS alone and in the presence of RNA. No significant difference is appreciated among all conditions. From left to right, n = 38, 39, 41, and 56. Kruskal–Wallis statistical test (n.s.: P-value > 0.05).

Figure 4.

Fluorescence lifetime as a sensor for the condensation degree of aS in a phase-separated system. (A) Box plots of the fluorescence lifetime (τ_L) of aS-atto488 in the dilute (left) and dense (right) phase at different RNA concentrations. From left to right, n = 19, 10, 14, 29, 28, 9, 9, and 29. One-way ANOVA and Kuskal–Wallis test followed by a Dunn’s test for the dilute phase (***P-value < 0.0005, ****P-value < 0.00005). (B) Confocal images of aS in LLPS buffer at different RNA concentrations. Top is LLPS buffer, bottom is after having diluted LLPS buffer twice with 20 mM PB.