Dimerisation and twist reversal of the Lewy fold in α-synuclein mutants with Parkinson's disease and dementia

This study reveals that pathogenic SNCA mutations (A53T, G51D) induce distinct left-handed Lewy fold dimer structures with twist reversal in familial Parkinson's disease and dementia, whereas the non-pathogenic H50Q variant and human A53T transgenic mouse models exhibit different filament architectures resembling wild-type or multiple system atrophy folds, respectively.

The Gist

The Big Picture: A Protein "Origami" Mystery

Imagine your brain is a bustling city, and inside every cell, there are millions of tiny workers called proteins. One specific worker, named Alpha-synuclein, is usually very helpful. It helps manage traffic and communication between cells.

However, sometimes this worker gets confused. Instead of staying in its normal shape, it folds up into a weird, rigid origami structure and clumps together with other confused workers. These clumps form long, stiff ropes called filaments. These ropes clog up the brain's machinery, leading to diseases like Parkinson's and Dementia.

For a long time, scientists thought all these ropes looked the same, regardless of what caused the disease. This paper, however, is like a detective story that reveals: "Wait a minute! The ropes look different depending on why the worker got confused."

---

The Three Characters: Different Mutations, Different Shapes

The researchers looked at the brains of people with specific genetic mutations (typos in the DNA instructions) that cause Parkinson's. They found three distinct "origami styles":

1. The "Twisted Double-Strand" (Mutations A53T and G51D)

• The Scenario: These mutations are like a "glitch" in the instructions that guarantees the disease will run in families.
• The Shape: Imagine a single rope (a filament) that usually twists to the right (like a standard screw). In these cases, the rope twists to the left. Even more surprisingly, instead of being a single rope, it's actually two ropes twisted together (a double helix).
• The Missing Piece: In the "normal" disease rope, there is a small, fuzzy patch of material attached to the side (called "Island A"). In these twisted double-ropes, this patch is missing.
• The Analogy: Think of a standard zipper. Usually, it has a plastic tab on the top (Island A). In these mutant cases, the tab is ripped off. Without the tab, the two sides of the zipper can zip together tightly, forming a double-strand. This double-strand is super stable and hard to break down, which makes the disease worse.
• The Result: Because the "tab" is missing, the rope twists left and doubles up. This seems to be the "toxic" version that causes severe, inherited Parkinson's.

2. The "Standard Single-Strand" (Variant H50Q)

• The Scenario: This is a mutation that looks scary on paper but turns out to be a "false alarm" in many cases.
• The Shape: These ropes look exactly like the ropes found in sporadic (random, non-inherited) Parkinson's. They are single strands, they twist to the right, and they keep the fuzzy patch (Island A) attached.
• The Analogy: This is like a standard zipper with the plastic tab still intact. Because the tab is there, the two sides can't zip together into a double-strand. They stay as single, less stable ropes.
• The Result: The researchers concluded that this specific mutation (H50Q) probably doesn't cause the disease on its own. The person likely had regular, random Parkinson's, and the mutation was just a coincidence.

3. The "Mouse Trap" (M83 Mice)

• The Scenario: Scientists have been using mice with the "A53T" mutation to study Parkinson's. They thought these mice were perfect models for human Parkinson's.
• The Surprise: When they looked at the ropes in the mice's brains, they didn't look like the human "A53T" ropes at all!
• The Shape: The mouse ropes looked like a completely different type of disease called MSA (Multiple System Atrophy). They were a unique shape that didn't match the human "twisted double-strand" at all.
• The Analogy: Imagine you are trying to fix a specific type of car engine (Human Parkinson's) by studying a toy car (The Mouse). You realize the toy car's engine is actually built like a completely different model of truck (MSA).
• The Lesson: We can't just use these mice to test drugs for human Parkinson's because their "clogs" are built differently. We need better models.

---

Why Does This Matter? (The "So What?")

Think of the brain as a house with a plumbing system.

• The "Right-Handed" Single Rope (H50Q/Idiopathic): These are like small clogs. They are annoying, but the water can still flow around them. They don't seem to be the main cause of the "inherited" disaster.
• The "Left-Handed" Double Rope (A53T/G51D): These are like a massive, welded steel pipe jamming the main water line. Because they are double-stranded and missing the "tab," they are incredibly strong and won't dissolve. This is the "gain of toxic function"—the mutation makes the protein too good at building these unbreakable clogs.

The Takeaway

1. Not all mutations are equal: Just because a gene has a typo doesn't mean it causes the disease. The H50Q mutation might be a "red herring."
2. Structure determines toxicity: The specific shape of the protein clump (single vs. double, left-twist vs. right-twist) dictates how bad the disease is.
3. Mouse models need a rethink: The mice we've been using to study Parkinson's might actually be modeling a different disease (MSA). We need to find ways to make mice that form the exact same "double-strand" ropes as humans do.

In short: This paper is like finding out that there are different types of "bad guys" in the brain. Some are single, weak thugs (H50Q), while others are a super-strong, double-banded gang (A53T/G51D). To stop the disease, we need to understand exactly how the gang forms so we can break them apart.

Technical Summary

1. Problem Statement

Parkinson's disease (PD), Parkinson's disease dementia (PDD), and dementia with Lewy bodies (DLB) are synucleinopathies characterized by the accumulation of $\alpha$-synuclein ($\alpha$-syn) filaments. While the structure of filaments in sporadic (idiopathic) PD has been solved (the "Lewy fold"), the structural impact of dominantly inherited SNCA mutations (specifically A53T, G51D, and H50Q) remained unknown.

• The Gap: Previous in vitro studies suggested that pathogenic mutations might induce entirely new filament folds distinct from the Lewy fold. However, it was unclear if filaments extracted directly from human brains with these mutations adopted the same fold as sporadic cases or if the mutations dictated a unique structural polymorph.
• The Model Discrepancy: Transgenic mouse models (e.g., M83 mice expressing human A53T) are widely used to study PD, but their filament structures had not been rigorously compared to human brain-derived filaments to validate their relevance.

2. Methodology

The study employed cryo-electron microscopy (cryo-EM) to determine high-resolution structures of $\alpha$-synuclein filaments extracted from post-mortem human brain tissue and transgenic mouse models.

• Sample Sources:
  • Human A53T: Temporal cortex from a patient with familial PDD (heterozygous A53T).
  • Human G51D: Frontal cortex from two patients with familial PDD (heterozygous G51D).
  • Human H50Q: Amygdala and frontal cortex from a patient with sporadic-like PD (heterozygous H50Q).
  • Mouse M83: Brainstems and spinal cords from homozygous M83 transgenic mice expressing human A53T.
• Extraction & Processing: Sarkosyl-insoluble fractions were extracted, and filaments were imaged using Titan Krios and G4 microscopes equipped with Falcon 4i and K3 detectors.
• Structure Determination: Helical reconstruction was performed using RELION. Atomic models were built using Coot and refined with Servalcat.
• Validation: Mass spectrometry (LC-MS/MS) was used to confirm the presence of specific mutations (A53T, G51D, H50Q) and species (mouse vs. human) within the filaments. Neuropathology (Gallyas-Braak silver staining) was performed to correlate structure with histological features.

3. Key Contributions & Results

A. Human A53T Mutation: Left-Handed Dimers without Island A

• Structure: Filaments exist as both singlets (10 nm) and doublets (20 nm).
• Twist: Unlike the right-handed twist of sporadic PD, A53T filaments exhibit a left-handed helical twist.
• Fold: The backbone conformation is essentially the Lewy fold (RMSD 1.42 Å vs. sporadic PD), but with a critical difference: Island A density is absent.
• Mechanism: The A53T mutation (Alanine $\to$ Threonine) disrupts the hydrophobic interface required for Island A binding. This loss of steric hindrance allows two protofilaments to dimerize. The dimer interface is formed by residues E46, V48, and H50.
• Implication: The mutation drives dimerization and a twist reversal (right $\to$ left) by eliminating the "Island A" cofactor density.

B. Human G51D Mutation: Left-Handed Dimers without Island A

• Structure: Only doublet filaments were observed (no singlets).
• Twist: Left-handed helical twist.
• Fold: Identical to the A53T doublet (RMSD 0.675 Å). The core is the Lewy fold, but Island A is missing.
• Mechanism: The G51D mutation (Glycine $\to$ Aspartate) introduces a negative charge, disrupting the hydrophobic interactions with Island A. This forces the protofilaments to dimerize via a shifted interface (E46, V48, H50) to stabilize the structure.
• Histology: Consistent with the structural dimerization, these inclusions were Gallyas-Braak silver-positive, unlike sporadic PD.

C. Human H50Q Variant: Right-Handed Singlets with Island A

• Structure: Filaments are singlets only.
• Twist: Right-handed helical twist.
• Fold: Nearly identical to the wild-type Lewy fold (RMSD 0.27 Å), including the presence of Island A and Island B.
• Significance: The H50Q variant does not disrupt the Island A interface. The structural similarity to sporadic PD, combined with the lack of family history in the patient, supports the hypothesis that H50Q is likely non-pathogenic and the patient suffered from idiopathic PD.

D. Mouse M83 Model: A Distinct "mA53T" Fold

• Structure: Filaments from homozygous M83 mice (expressing human A53T) formed a completely different fold (named mA53T fold).
• Comparison: This fold is not the Lewy fold. It has an Amyloid Packing Difference (APD) of 76% compared to human Lewy folds.
• Similarity: The mA53T fold closely resembles the MSA (Multiple System Atrophy) fold (APD 38%) and the fold found in juvenile-onset synucleinopathy.
• Conclusion: Despite expressing the human A53T mutation, the M83 mouse model generates filaments structurally related to MSA, not PD. This explains why M83 mice are better models for MSA than for PD.

4. Significance and Mechanistic Insights

• Pathogenicity Mechanism: The study establishes a direct link between protofilament dimerization, left-handed twist, and dominant inheritance in PD.
  • Gain-of-Toxic-Function: Pathogenic mutations (A53T, G51D) disrupt the binding of a cofactor (Island A). This loss of steric bulk allows protofilaments to dimerize.
  • Stability: Dimeric filaments are likely more stable due to increased hydrogen bonding and avidity, leading to persistence in cells and neurodegeneration.
  • Silver Staining: The dimeric, left-handed structure correlates with positive Gallyas-Braak silver staining, explaining why familial cases stain differently than sporadic cases.
• Re-evaluation of Mouse Models: The finding that M83 mice produce MSA-like folds rather than PD-like Lewy folds challenges the use of this specific line for modeling familial PD. It suggests that the cellular environment (mouse vs. human) or the specific seeding mechanism dictates the fold, overriding the mutation's effect seen in humans.
• Therapeutic Implications:
  • Understanding that familial PD involves a specific dimeric, left-handed Lewy fold (lacking Island A) provides a precise target for structure-based drug design.
  • Current in vitro assembly methods fail to reproduce the human brain Lewy fold, highlighting a critical gap in developing mechanism-based therapies.

Summary Table of Structures

Sample	Mutation	Filament Type	Handedness	Island A Present?	Fold Type	Silver Staining
Sporadic PD	WT	Singlet	Right	Yes	Lewy	Negative
Human A53T	A53T	Singlet & Doublet	Left	No	Lewy (Dimeric)	Positive
Human G51D	G51D	Doublet	Left	No	Lewy (Dimeric)	Positive
Human H50Q	H50Q	Singlet	Right	Yes	Lewy	Negative
M83 Mouse	Human A53T	Singlet & Doublet	Left	No	mA53T (MSA-like)	N/A

This paper fundamentally shifts the understanding of familial synucleinopathies, demonstrating that specific mutations drive a structural switch from monomeric right-handed filaments to dimeric left-handed filaments, a process that is distinct from the structures observed in current mouse models.