A crawling cell extends a thin, sheet-like lamellipodium at its leading edge. Inside it, a dense meshwork of actin filaments grows by adding subunits at their "barbed" (plus) ends pushed against the plasma membrane, while older filament segments are disassembled farther back — a process called treadmilling. The network itself flows backward relative to the substrate (retrograde flow) even as the membrane edge advances, and net forward motion only happens once focal adhesions grip the substrate and convert that polymerization force into traction.
The "molecular clutch" hypothesis explains why cells crawl fastest at intermediate adhesion strength — too little grip and the actin just slips backward (pure retrograde flow), too much grip and filaments buckle under load, but a moderate clutch lets polymerization force couple efficiently to substrate traction.
An interactive 3D leading edge where actin filaments treadmill — growing at the membrane and disassembling farther back — while focal adhesions grip the substrate and convert that polymerization force into forward crawling.
Net crawl speed depends on the balance between actin polymerization pushing the membrane forward and retrograde flow pulling the network backward, mediated by how strongly focal adhesions "clutch" the substrate.
Raise polymerization rate to push the leading edge harder, tune adhesion strength to see the molecular clutch engage or slip, and toggle Arp2/3 branching to compare a dense dendritic meshwork against sparse, unbranched filaments.
Cells often crawl fastest at intermediate adhesion strength — too little grip wastes polymerization force as pure retrograde flow, too much grip stalls the network under load.