β-motifs and molecular flux promote amyloid nucleation at condensate interfaces

This paper introduces the Flux-Driven Molecular Dynamics (FD-MD) framework to demonstrate that biomolecular condensate interfaces significantly accelerate amyloid nucleation by reducing entropic costs, while molecular flux and segment rigidity jointly dictate non-equilibrium growth morphologies and aging trajectories.

The Gist

The Big Picture: From Liquid Droplets to Hard Sticks

Imagine your cells are filled with tiny, floating liquid droplets (called biomolecular condensates). Think of these like oil droplets in a salad dressing or water droplets on a foggy window. Inside these droplets, proteins float around freely, mixing and matching to get work done. This is the healthy, "liquid" state.

However, sometimes these droplets get old and "sick." The proteins inside them stop flowing and start clumping together into hard, rigid, stick-like structures called amyloid fibrils. This is the "solid" state, and it's a hallmark of diseases like Alzheimer's and ALS.

The big mystery scientists have been trying to solve is: How does a smooth, liquid droplet suddenly turn into a hard, spiky solid?

This paper introduces a new computer simulation method (called FD-MD) to watch this process happen in slow motion. They discovered that the answer lies in two main things: the shape of the protein and how fast new proteins are being delivered.

---

1. The "Velcro" Effect: Why the Edge is Special

Imagine you have a room full of people (the proteins) wearing different outfits. Some are wearing floppy pajamas (flexible proteins), and some are wearing stiff, rigid suits with Velcro patches on them (rigid, $\beta$-prone segments).

• In the middle of the room (the bulk): If two people in stiff suits try to hold hands, they have to spin around and find each other in 3D space. It's hard to coordinate; they might bump into others first.
• At the edge of the room (the interface): Now, imagine these people are standing against a wall. The wall stops them from spinning around in all directions. They are forced to face the same way. Suddenly, it's much easier for the Velcro patches to find each other and stick together.

The Finding: The paper shows that the surface (interface) of the droplet acts like that wall. It reduces the "confusion" (entropy) of the proteins, making it 100 times easier for them to line up and start building a solid structure right at the edge of the droplet, rather than in the middle.

2. The "Construction Site" Analogy: Two Ways to Build

The researchers simulated a construction site where new workers (proteins) are constantly arriving to build a structure on the droplet's surface. They found that the speed of delivery changes the final building completely.

Scenario A: The Slow Delivery (Low Flux)

Imagine a construction crew where workers arrive one by one, slowly.

• What happens: Each worker has time to walk to the very tip of the growing structure and attach themselves perfectly.
• The Result: They build long, thin, needle-like fibrils that shoot out from the droplet. It's like a slow-growing crystal.
• Analogy: It's like a slow-growing stalactite in a cave. Water drips slowly, allowing minerals to crystallize into long, thin spikes.

Scenario B: The Fast Delivery (High Flux)

Imagine a firehose of workers being sprayed onto the site all at once.

• What happens: The workers arrive so fast that they don't have time to walk to the tips. They just pile up on the flat surface where they land.
• The Result: Instead of long needles, you get a thick, flat, messy layer covering the droplet. The "needles" never get a chance to grow because the surface gets saturated too quickly.
• Analogy: It's like trying to build a tower of bricks while someone is dumping a truckload of bricks on your head every second. You can't build up; you just get a big, flat pile.

3. The "Traffic Jam" of Movement

The paper also looked at how the proteins move once they are stuck.

• Healthy State: Proteins are like fish swimming in a pond. They can move anywhere.
• Sick State: Once the proteins lock into those rigid, Velcro-like structures, they get stuck. It's like the fish are now trapped in a cage made of ice. They can't move anymore. The computer simulation showed that as the "solid" structure grows, the movement of the proteins slows down drastically, eventually stopping completely. This "freezing" is a sign that the droplet has turned into a solid, pathological mass.

4. The "Blueprint" Matters

The researchers also tested what happens if the "Velcro patches" are scattered randomly along the protein chain instead of being grouped together.

• Result: Nothing happens. The proteins pile up, but they don't form the hard needles.
• Lesson: It's not just about having the "sticky" parts; it's about where they are located. The proteins need a specific "blueprint" (sequence) where the sticky parts are grouped together to trigger the hardening.

The Takeaway: A New Way to Think About Disease

This paper suggests that the "aging" of these cellular droplets isn't just about the proteins themselves being bad. It's a traffic problem.

1. The Blueprint: The protein must have the right shape (rigid, sticky patches).
2. The Traffic: The cell must be sending too many proteins to the droplet too fast (or too slow).

Why does this matter?
If we can understand that the speed of protein delivery controls whether a droplet stays liquid or turns into a solid, we might be able to treat diseases without changing the protein's DNA. Instead of fixing the "broken" protein, we could just slow down the traffic (reduce the supply rate) to keep the droplets liquid and prevent them from turning into the hard, disease-causing solids.

In short: The edge of the droplet is the danger zone where hardening starts, and the speed of the supply line determines whether you get long, dangerous needles or just a harmless, flat pile.

Technical Summary

1. Problem Statement

Biomolecular condensates (membraneless organelles) are increasingly recognized as intermediates in the formation of pathological amyloid assemblies associated with neurodegenerative diseases. While experimental evidence suggests that condensate interfaces act as preferential sites for fibril nucleation, the underlying kinetic mechanisms remain poorly understood. Specifically, there is a gap in understanding how sequence-encoded structural motifs (such as rigid $\beta$-prone segments) and non-equilibrium molecular transport (driven influx) cooperate to drive the transition from liquid-like condensates to solid-like fibrillar states.

Existing computational methods face limitations:

• Atomistic simulations: Capture sequence-specific interactions but are limited to timescales and system sizes too small to observe interface-mediated fibril growth.
• Coarse-grained models: Can access mesoscale dynamics but often lack explicit representations of $\beta$-prone motifs or sustained non-equilibrium molecular exchange.

2. Methodology: Flux-Driven Molecular Dynamics (FD-MD)

To address these limitations, the authors introduce Flux-Driven Molecular Dynamics (FD-MD), a non-equilibrium simulation framework designed to couple structural determinants with kinetic transport.

• Molecular Model: Protein chains are modeled as coarse-grained polymers consisting of:
  • Rigid, $\beta$-prone segments: Modeled with harmonic angular potentials to enforce rod-like geometry and attractive Lennard-Jones interactions to mimic $\beta$-sheet cohesion.
  • Flexible linkers: Representing intrinsically disordered regions.
• Simulation Setup:
  • Two planar condensate slabs act as interfaces within a simulation box.
  • Non-equilibrium Influx: Polymer chains are continuously inserted into the dilute phase at a controlled rate ($\dot{m}$) with an initial drift velocity ($v_0$) directed toward the interfaces. This mimics a chemical potential gradient driving molecular transport.
  • Thermostat: A Nosé–Hoover thermostat maintains constant temperature while allowing the system to remain open and non-equilibrium.
• Key Parameters: The study systematically varies:
  • Segment Stiffness ($\ell_p$): Quantified by persistence length.
  • Supply Rate ($\dot{m}$): The rate of molecular delivery.
  • Drift Velocity ($v_0$): The speed of transport toward the interface.

3. Key Contributions

1. Development of FD-MD: A novel simulation framework that simultaneously integrates sequence-encoded $\beta$-forming interactions, interfacial anisotropy, and persistent non-equilibrium influx.
2. Entropic Scaling Analysis: A theoretical derivation showing how condensate interfaces reduce the entropic cost of aligning rigid segments, thereby lowering the nucleation barrier.
3. Non-Equilibrium Phase Diagram: The mapping of four distinct interfacial growth morphologies based on stiffness and supply rate.
4. Kinetic Mechanisms: Identification of the competition between planar surface saturation and directed tip elongation, governed by transport rates.

4. Key Results

A. Interface-Promoted Nucleation via Entropic Reduction

• Mechanism: In the bulk dilute phase, rigid segments explore a full solid angle ($4\pi$). At a flat interface, geometric exclusion restricts this to a hemisphere ($2\pi$).
• Result: This restriction lowers the orientational entropy cost of co-alignment by $\Delta\Delta S \approx (n-1)k_B \ln 2$ per segment pair.
• Impact: For a critical nucleus size of 2 chains, this entropic advantage enhances the interfacial nucleation rate ($J_{surf}$) relative to the bulk ($J_{bulk}$) by approximately two orders of magnitude ($J_{surf}/J_{bulk} \approx 256$).

B. Non-Equilibrium Phase Diagram

Varying segment stiffness ($\ell_p$) and supply rate ($\dot{m}$) reveals four distinct growth regimes:

1. Uniform Wetting (Low Stiffness): Fully flexible chains lead to isotropic, planar thickening of the interface without fibril formation.
2. Disordered Deposition (Low-Moderate Stiffness): Aggregates accumulate without long-range order, forming heterogeneous, "craze-like" surfaces.
3. Fibrillar Interfacial Growth (Moderate-High Stiffness): Localized rigid segments nucleate surface-anchored protofilaments that elongate into fibrillar protrusions.
4. Bridging Networks (High Stiffness + High Supply): Fibrils bundle and connect opposing condensate interfaces, forming mechanically stiff, load-bearing networks.

Crucially, the spatial organization of rigid motifs is essential. Randomly distributed rigid segments fail to nucleate fibrils, producing disordered deposits, confirming that sequence patterning dictates the aging pathway.

C. Kinetics of Surface Growth and Transport Limits

The study identifies a crossover between two kinetic regimes based on drift velocity ($v_0$):

• Synthesis-Limited Regime (Low $v_0$): The supply rate ($\dot{m}$) dominates. Increasing $\dot{m}$ accelerates fibril nucleation and shifts the system to a linear growth regime ($\rho_{surf} \sim t$) earlier.
• Transport-Limited Regime (High $v_0$): Chains arrive faster than they can be incorporated into fibril tips. The system becomes saturated, and increasing $\dot{m}$ has diminishing returns on growth rate.
• Growth Law: The effective growth rate follows a hyperbolic saturating function: $R(|v_0|) = \frac{R_{max} \dot{m} |v_0|}{v^*_0 + |v_0|}$.

D. Directional Elongation vs. Drift Velocity

• Inverse Relationship: The longitudinal elongation velocity ($v_{elong}$) of fibrils is inversely proportional to the drift velocity ($v_0$).
  • Mechanism: At high $v_0$, incoming chains rapidly saturate the flat interface (planar coverage) before they can adopt the specific orientation required for tip incorporation. This diverts material away from fibril tips, slowing elongation.
  • Low $v_0$: Allows sufficient time for chains to find and dock at growing filament tips, sustaining faster directional elongation.
• Radial Growth: In contrast, radial thickening ($R_\perp$) is largely insensitive to $v_0$, suggesting it is driven by local diffusion-limited attachment rather than global transport.

E. Dynamic Arrest

• MSD Analysis: Chains incorporated into fibrillar structures exhibit a transition from superdiffusive (ballistic) motion to subdiffusive dynamics ($\langle \Delta r^2 \rangle \sim t^{0.1}$), indicating dynamic arrest.
• Significance: This subdiffusive behavior serves as a molecular-level marker for the maturation of condensates from liquid-like to solid-like states. Rigid segments are essential for inducing this arrest; flexible chains remain diffusive.

5. Significance and Implications

• Mechanistic Framework: The paper provides a unified picture where condensate aging is determined by two independent axes: sequence-encoded rigidity (structural competence) and molecular flux (kinetic selection).
• Therapeutic Potential: The findings suggest that aging trajectories can be redirected without altering the protein sequence itself. Modulating molecular supply (e.g., via protein expression levels or intracellular transport rates) could potentially prevent the transition to pathological fibrillar states.
• General Applicability: The FD-MD framework offers a general tool for studying how non-equilibrium mass transport interacts with sequence features to drive mesoscale organization, applicable beyond amyloid systems to any condensate undergoing aging.
• Experimental Predictions: The model predicts that the aspect ratio of fibrillar products is tunable via transport conditions (slower delivery $\rightarrow$ longer, thinner fibrils; faster delivery $\rightarrow$ shorter, thicker deposits) and that subdiffusive dynamics can serve as an experimental readout for fibrillar maturation.