Direct Observation of β-Barrel Intermediates in the Self-Assembly of Toxic SOD128-38 and Absence in Nontoxic Glycine Mutants

Abstract

Soluble low-molecular-weight oligomers formed during the early stage of amyloid aggregation are considered the major toxic species in amyloidosis. The structure-function relationship between oligomeric assemblies and the cytotoxicity in amyloid diseases are still elusive due to the heterogeneous and transient nature of these aggregation intermediates. To uncover the structural characteristics of toxic oligomeric intermediates, we compared the self-assembly dynamics and structures of SOD1_28-38, a cytotoxic fragment of the superoxide dismutase 1 (SOD1) associated with the amyotrophic lateral sclerosis, with its two nontoxic mutants G33V and G33W using molecular dynamics simulations. Single-point glycine substitutions in SOD1_28-38 have been reported to abolish the amyloid toxicity. Our simulation results showed that the toxic SOD1_28-38 and its nontoxic mutants followed different aggregation pathways featuring distinct aggregation intermediates. Specifically, wild-type SOD1_28-38 initially self-assembled into random-coil-rich oligomers, among which fibrillar aggregates composed of well-defined curved single-layer β-sheets were nucleated via coil-to-sheet conversions and the formation of β-barrels as intermediates. In contrast, the nontoxic G33V/G33W mutants readily assembled into small β-sheet-rich oligomers and then coagulated with each other into cross-β fibrils formed by two-layer β-sheets without forming β-barrels as the intermediates. The direct observation of β-barrel oligomers during the assembly of toxic SOD1_28-38 fragments but not the nontoxic glycine-substitution mutants strongly supports β-barrels as the toxic oligomers in amyloidosis, probably via interactions with the cell membrane and forming amyloid pores. With well-defined structures, the β-barrel might serve as a novel therapeutic target against amyloid-related diseases.

Fig. 1.

Secondary structure analyses of SOD1_28–38, SOD1_28–38^G33V and SOD1_28–38^G33W self-assemblies, (a-c) The average secondary structure contents in terms of coil, β-sheet, helix, bend and turn for SOD1_28–38, SOD1_28–38^G33V and SOD1_28–38^G33W aggregation with different number of peptides. The propensity of each residue adopted coil (d-f), β-sheet (g-i), bend (j-k) and turn (m-o) for each type of peptides with different system sizes. The error bars of secondary structure propensities correspond to the standard deviations of means from 20 independent simulations. Our results suggest that substituting flexible glycine at residue position 33 (highlighted in red in panels d-o) with hydrophobic valine and tryptophan significantly increased the β-sheet propensity in the aggregates.

Fig. 2.

Conformational analyses of the SOD1_28–38, SOD1_28–38^G33V and SOD1_28–38^G33W assemblies. The probability distribution of the oligomer size (a-c), β-sheet oligomer size (d-f) and β-sheet size (g-i) for the SOD1_28–38, SOD1_28–38^G33V and SOD1_28–38^G33W self-assemblies in each molecular system. To avoid bias from the initial state, only the conformations of the last 200 ns of all independent simulations were included in the analysis. For each peptide type, representative aggregate structures in simulations of 16 peptides were also presented (j-k). SOD1_28–38, SOD1_28–38^G33V and SOD1_28–38^G33W peptides were colored in cyan, purple and pink, respectively.

Fig. 3.

The inter–peptide interaction analysis. The residue-pairwise inter-molecular contact frequency maps were computed both between main-chain atoms (MC-MC) and between side-chain atoms (SC-SC) in simulations of sixteen peptides. Only the last 200ns of each 1000ns independent simulations was used for analysis.

Fig. 4.

The self-assembly dynamics and conformations analysis. The self-assembly dynamics of eight SOD1_28–38 a), SOD1_28–38^G33V b), and SOD1_28–38^G33W c) peptides monitored by the time evolution of the largest oligomer size (black), largest β-sheet oligomer size (red), largest β-sheet size (blue), averaged β-sheet layer size (purple) and β-barrel size (green). The oligomer structures along the simulation trajectory as indicated by cyan arrows were present at the bottom, d) The probability of observing the formation of β-barrels in each molecular system for all three peptide types, e) The probability distribution of β-barrels with different sizes was calculated as a function of the number of simulated peptides for SOD1_28–38. f-g) The representative structure of SOD1_28–38 β-barrel pentamer and hexamer. SOD1_28–38, SOD1_28–38^G33V and SOD1_28–38^G33W were colored in cyan, purple and pink, respectively.

Fig. 5.

The self-assembly free energy landscape and aggregation dynamics analysis for sixteen-peptide simulations. For each of three peptide types - a) SOD1_28–38, b) SOD1_28–38^G33V and c) SOD1_28–38^G33W, the free energy landscape as a function of the oligomer size and the average number of residues adopting β-sheet conformation per chain (N_β-sheet) are presented on the left. To capture all the conformation of assemblies during the aggregation process, the whole 1000 ns trajectory of 20 independent DMD runs were included in the analysis. A representative trajectory from isolated monomers into final nano-fibrils via all the intermediates corresponding to labeled basins in the PMF is also shown on the right. The size of the largest oligomer (black), largest β-sheet oligomer (red), largest β-sheet layer (blue), averaged β-sheet layer (purple) and β-barrel (green) are plotted as a function of simulation time. The snapshots along with the growth of assemblies are also shown in the inset on the right and their corresponding states in the free energy landscape are also labeled. SOD1_28–38, SOD1_28–38^G33V and SOD1_28–38^G33W peptides are colored in cyan, purple and pink, respectively.