⚛️ biophysics

Mechanosensitive Remodeling Sustains Rigidity Homeostasis and Elastic Memory in Actin Cortex Models

This paper proposes two complementary elastic network models demonstrating that mechanosensitive remodeling dynamics, characterized by preferential disassembly under low tension and energy-injected assembly, enable actin cortex networks to maintain rigid homeostasis and long-lived elastic memory despite continuous structural turnover.

Original authors: Wang, H., Galvani Cunha, M. A., Crocker, J. C., Liu, A. J.

Published 2026-07-13

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Wang, H., Galvani Cunha, M. A., Crocker, J. C., Liu, A. J.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine a city made entirely of stretchy rubber bands and tiny, invisible magnets. This city is the actin cortex, a dynamic skin just under the surface of your cells. It has a superpower that seems impossible: it is constantly being torn down and rebuilt, yet it stays stiff and remembers how it was stretched long after the original pieces have vanished.

Usually, if you keep replacing the bricks in a wall, the wall would fall apart. But the actin cortex is different. It's like a city where the buildings are constantly being demolished and rebuilt, yet the city's layout somehow stays rigid, and if you push the city, it "remembers" that push for a long time, even though the specific buildings that felt the push are long gone.

The Secret Sauce: A Smart, Self-Healing City

The authors of this paper built two computer models to figure out how this magic trick works. They treated the cell's skeleton like a network of springs (the rubber bands) and nodes (the magnets). They discovered that for this city to stay rigid while constantly changing, it needs three specific rules, or "ingredients":

The "Weak Link" Rule: If a rubber band (filament) or a magnet (crosslinker) isn't being pulled very hard, it gets removed. It's like a city that only keeps the buildings that are currently under stress and tears down the ones that are just hanging out doing nothing.
The "Random Chaos" Rule: Sometimes, a piece gets removed just by chance, even if it's under tension. This ensures the city never gets stuck in a loop and keeps changing.
The "Energy Boost" Rule: When a new piece is added, it's installed with a little bit of a stretch (prestrain). This injects energy back into the system to replace what was lost.

The paper suggests that without all three of these rules working together, the city fails. If you only have random removal, the city loses its stiffness. If you only remove weak links without adding energy, the city collapses. It's a delicate balance.

The "Tunable Matter" Idea

The authors propose that the actin cortex is a type of "tunable matter." Think of it like a giant, living neural network for a robot. In a computer, you tune the numbers inside a program to make it smart. In the cell, the "tuning" happens physically: the springs and magnets adjust themselves based on how much force they feel.

The paper explicitly argues against the idea that the cell just needs a static, unchanging structure to stay rigid. They show that a network with too few connections (which is what the cell actually has) would normally be floppy and fall apart unless it uses these smart, force-sensitive rules to constantly reorganize itself.

The "Ghost" Memory

Here is the wildest part: Elastic Memory.

When you stretch a normal rubber band and let go, it snaps back quickly. But the actin cortex is different. If you stretch it, the stress (the tension) doesn't fade away quickly like a normal exponential decay. Instead, it fades away very slowly, following a mathematical pattern called a power law.

The paper suggests this happens because the "smart rules" preserve the pattern of tension. Even though the specific rubber bands that were stretched at the beginning are gone (replaced by new ones), the new ones are arranged in a way that keeps the same "tension map." It's like a game of telephone where the message changes slightly every time, but the core story remains the same for a long time.

The "Drifting" City

In these simulations, the pieces of the city (the nodes and edges) don't just sit still; they drift around. The paper notes that the movement of these pieces is "heavy-tailed," meaning sometimes they take tiny steps, but other times they take huge jumps.

This leads to a phenomenon called "representational drift." The city's function (staying rigid) stays the same, but the microscopic details (which specific springs are where) are constantly changing. It's like a band playing the same song, but every night, the musicians swap instruments and positions, yet the song sounds perfect.

How Sure Are We?

It is important to note that these results come from computer simulations, not direct measurements of a living cell in a petri dish. The authors built these models to test if their "ingredients" could produce the behavior we see in real cells.

What they proved in the simulation: They showed that if you program a network with these three rules, it does achieve rigidity, complete turnover (everything gets replaced), and long-lasting memory.
What they suggest: They suggest that real cells likely use these exact mechanisms. They point to real biological experiments that show weak links are indeed removed faster and that crosslinkers act like "catch bonds" (sticking tighter under certain forces), which supports their model.
What they ruled out: They showed that purely random removal or purely force-sensitive removal without the other ingredients fails to maintain rigidity.

The paper also mentions that these simulated networks are surprisingly close to a "critical point"—a state where the network is on the edge of being floppy. This suggests the cell might be fine-tuned to be just stiff enough to hold its shape but flexible enough to adapt, a state known as "self-organized criticality."

The Bottom Line

The actin cortex isn't a static scaffold; it's a self-healing, self-tuning machine. By constantly tearing down weak parts and rebuilding them with a little extra stretch, it maintains a rigid shape and remembers its history, all while its individual parts are in a state of constant, fluid motion. The paper suggests this is how life manages to be both strong and adaptable at the same time.

Technical Summary: Mechanosensitive Remodeling Sustains Rigidity Homeostasis and Elastic Memory in Actin Cortex Models

Problem Statement
The actin cortex is a dynamic biopolymer network that maintains a rigid homeostatic state (positive elastic moduli) and exhibits long-lived elastic memory despite constant structural turnover. In biological systems, filaments and crosslinkers undergo rapid assembly and disassembly, with unchanged components decaying exponentially on timescales of tens of seconds. This creates a paradox: how can a network maintain rigidity and "remember" a deformed state when its structural constituents are completely replaced? Conventional mechanical intuition suggests that rigidity should be lost if the network is undercoordinated (mean coordination number $\bar{z} < 6$ ) and subject to random remodeling. Furthermore, while hyperstatic networks ( $\bar{z} > 6$ ) are inherently rigid, many cell types operate with $\bar{z} \approx 3\text{--}4$ , relying on filament tension (tensegrity) and bending stiffness. However, neglecting bending stiffness to isolate the tension mechanism reveals that random pruning alone causes rapid loss of rigidity. The fundamental question addressed is how the cortex achieves robust rigidity and elastic memory through complete turnover.

Methodology
The authors develop two complementary minimal elastic network models to simulate the actin cortex as a "tunable matter" system, where local mechanosensitive rules adjust the network's degrees of freedom (edges or nodes) to achieve collective rigidity. The models consist of 3D harmonic spring networks (representing filaments) joined by point-like nodes (representing crosslinkers/branchers) under periodic boundary conditions. Bending rigidity is neglected to focus on stretching-dominated mechanics.

The two models differ in which element is dynamic:

Mechanosensitive Edge Model: Nodes are persistent; edges (filaments) are removed or inserted.
Mechanosensitive Node Model: Edges are persistent; nodes (crosslinkers) split or merge, changing connectivity.

Both models incorporate three minimal ingredients derived from biological observations and theoretical necessity:

Preferential Disassembly: Edges under low tension (Edge model) or nodes exerting small forces (Node model) are preferentially removed. This mimics tension-inhibited pruning and catch-bond behavior.
Random Disassembly: A small, non-zero rate of random removal ( $\kappa$ ) is included to ensure complete turnover.
Energy Injection: Upon assembly (edge insertion or node merging), a prestrain $\epsilon$ is imposed, injecting elastic energy to compensate for dissipation.

Simulations employ a quasi-static limit where force equilibration (Newton-Raphson minimization) occurs much faster than structural turnover. The models are initialized from jammed packing states with $\bar{z} \approx 3.32$ and evolved to steady states.

Key Contributions and Results

Rigidity Homeostasis with Complete Turnover: Both models successfully reach a dynamic steady state where the bulk modulus ( $B$ ) and shear modulus ( $G$ ) remain positive and stable, despite the continuous replacement of all structural elements. The turnover of edges or nodes follows an exponential decay, with characteristic timescales ( $\tau_E \approx 1335$ iterations, $\tau_V \approx 450$ iterations) significantly shorter than the time to reach steady state.
Necessity of Minimal Ingredients: Sensitivity analysis demonstrates that all three ingredients are necessary and sufficient.
- Without random disassembly, the network fails to turn over completely (stalling at ~93% unchanged elements).
- Without preferential (mechanosensitive) disassembly, the network loses rigidity as elastic energy decays to zero.
- Without energy injection, rigidity cannot be sustained.
Emergence of Elastic Memory: Upon applying a step strain, the stress relaxation modulus $G(t)$ decays algebraically ( $t^{-1}$ ) rather than exponentially. This long-lived memory persists well beyond the structural turnover time (retaining 10–20% of initial stress after $10\tau$ ). The origin of this memory is identified as the preservation of tension-weighted anisotropy: edges aligned with the stretch direction experience higher tension and are thus less likely to be disassembled, maintaining the anisotropic structure responsible for the stress.
Diffusive Dynamics and Representational Drift: Despite maintaining solid-like rigidity, the nodes and edges exhibit diffusive motion with mean-squared displacements (MSD) growing linearly with time, exceeding the Lindemann criterion for solids. The displacement distributions are heavy-tailed (fitted by truncated Lorentzians). This behavior, combined with complete turnover, constitutes "representational drift," where the microscopic encoding of macroscopic rigidity constantly shifts without permanent structural correlations.
Robustness and Self-Healing: The models are robust to variations in initial conditions and random disassembly rates. Crucially, networks can recover from severe structural disruption (e.g., a large hole representing 25% of the volume) within $\sim 2\tau$ , restoring elastic moduli to steady-state values.
Proximity to Criticality: The steady-state networks operate near the rigidity phase transition (critical point). The shear modulus drops sharply under small compressional strains, suggesting the cortex functions in a self-organized critical (SOC) state, balancing between floppy and rigid phases.

Significance
The paper proposes that the actin cortex functions as "tunable matter," utilizing mechanosensitive local rules to explore a vast space of network architectures. Unlike engineered materials or standard neural networks that rely on global optimization or static structures, the cortex achieves robust rigidity, structural healing, and elastic memory through local, biologically plausible mechanisms.

The authors argue that this framework resolves the paradox of how a rapidly remodeling, undercoordinated network maintains mechanical integrity. The "tunable" nature of the network—where edges and nodes are individually adjustable based on local tension—allows the system to maintain robust but flexible rigidity in fluctuating mechanical environments. The work formalizes the intuition that turnover enables adaptability, providing a quantitative mechanism where the cortex uses local rules to tune its degrees of freedom to achieve biologically desirable properties. This perspective contrasts with the "use it or lose it" principle, suggesting instead that the cortex actively tunes its architecture to perform mechanical functions, exhibiting behaviors analogous to physical learning circuits and representational drift found in neural systems.