Computer Simulations of Mechano-Chemical Networks Choreographing Actin Dynamics in Cell Motility - Computational Modeling of Biological Systems - page 248

Fig. 3 The following processes were included into our computational model of a lamellipodial
protrusion [ 80 ]: (1) Stochastic hopping of various monomeric proteins between neighboring
voxels; (2) polymerization and depolymerization events for individual F-actin filaments; (3)
binding of Arp2/3 to sides of existing filament to nucleate a daughter filament at approximate 70 ı
angle; (4) binding of capping proteins and formins to polymer barbed ends, to correspondingly
stop and accelerate the polymerization process. Filaments sterically protrude against the cell
membrane, locally deforming it, while the membrane provides some resistance against the bending
deformations and an increase of the membrane area. The counteracting membrane push against
the filaments slows down polymerization of filament tips that bear the most force. A boundary
with bulk reservoir of monomers is placed at the back of the lamellipodium, several microns
down in the x direction. This results in establishing of monomeric gradients longitudinally across
the lamellipodium for species that are actively consumed in front, and have to be diffusively
transported from the rear to replenish the local pool
concentration, filamentous network is sparse, leading to high protrusive resistance
on filaments; on the other hand, at high Arp2/3 concentration, actin filaments
in the dense network would deplete the local monomeric actin pool. Both cases
are unfavorable to the polymerization of actin filaments, although the causes are
different. Overall, this observation indicates that having balanced polymerization
and nucleation rates is important in order to producing maximal protrusion speeds.
The results obtained from microscopic simulations [ 35 ] are qualitatively consistent
with the theoretical analysis using a set of deterministic reaction-diffusion partial
differential equations [ 66 ].