Cell Motility: Actin Polymerization & Crawling

Crawling without muscles or motors

A white blood cell chasing a bacterium, or a fibroblast healing a wound, does not swim and has no legs — it crawls, and the engine behind that crawl is not a muscle motor at all but the constant, directional assembly and disassembly of a single protein: actin. At the cell's leading edge, actin monomers polymerize into filaments fast enough to physically push the cell membrane forward, and this simulation renders that process directly — a dense, ever-renewing meshwork of actin filaments growing at the front of a crawling cell.

Treadmilling: growing at one end, shrinking at the other

Actin filaments are polar — structurally different at their two ends, called barbed (plus) and pointed (minus). Free actin monomers (G-actin, bound to ATP) preferentially add onto the barbed end; at the pointed end, older subunits (whose bound ATP has since hydrolyzed to ADP) preferentially fall off. When the addition rate at the barbed end matches the loss rate at the pointed end, the filament's overall length stays constant even though every individual subunit is migrating from one end of the filament to the other — a process called treadmilling. It costs one ATP hydrolysis per monomer cycled through, which is the actin cytoskeleton's real energy expenditure for crawling.

barbed (+) end:  fast ATP-actin monomer addition   (growth)
pointed (-) end: slow ADP-actin monomer loss        (disassembly)

net filament length ~ constant when addition rate ~ loss rate,
but individual subunits flow continuously from + end to - end

The Brownian ratchet: how growth becomes a push

Polymerization alone does not obviously explain how a filament pushes a membrane — a rigid rod growing against a wall would just stop growing, blocked. The resolution is the Brownian ratchet model (Peskin, Odell and Oster, 1993): the membrane is not static, it is being jostled by thermal (Brownian) motion, flexing back and forth by nanometres on its own. Most of the time this gap is too small for another actin monomer to slot onto the filament tip. But whenever a random thermal fluctuation opens a gap just larger than the ~2.7 nm added by one monomer, a new subunit can add before the membrane flexes back — and once added, that subunit is a rigid strut preventing the membrane from returning as far as it started. The membrane is effectively ratcheted forward, one monomer-width at a time, by rectifying its own random thermal motion rather than by the filament exerting a mechanical push in the conventional sense.

Branching: the Arp2/3 complex and the dendritic network

A single filament growing by Brownian ratcheting produces a weak, easily-buckled force. Real lamellipodia (the flat, sheet-like leading edge of a crawling cell) instead build a dense, self-reinforcing meshwork using the Arp2/3 complex, which binds to the side of an existing actin filament and nucleates a new filament branching off it at a characteristic 70 degree angle. Each new branch is itself a template for further branching, producing a treadmilling dendritic network — a name borrowed directly from its resemblance to a branching tree — dense enough that many filament tips push on the membrane simultaneously, distributing the load and making the whole front edge mechanically stiff enough to advance coherently rather than buckle.

Anchoring the push: focal adhesions

A ratchet that pushes the membrane forward accomplishes nothing for whole-cell translocation unless the rest of the cell has something to brace against — otherwise the cell would simply push its front forward while its back slides back an equal amount, no net progress. Focal adhesions are the anchor: transmembrane integrin proteins that bind the actin network on the inside to the extracellular matrix on the outside, at discrete points near the leading edge. With the front edge's actin network gripped to the substrate, the polymerization-driven push at the tip is converted into forward translocation of the whole cell body rather than a purely local membrane ripple. Focal adhesions at the trailing edge must then release for the rear of the cell to be pulled forward, coordinated by myosin-II-generated tension along the actin network deeper inside the cell — crawling is a cycle of protrude-at-front, adhere, contract, release-at-back, repeated continuously.

Frequently asked questions

Try it live

Everything above runs in your browser — open Cell Motility and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.