Use this permanent link to cite or share this Morpheus model:
Cell motility simulation with a reaction-diffusion model for actin waves on the cell perimeter
Introduction
A partial differential equation model for ‘actin waves’ by Holmes et al., 2012 was simplified and investigated by PDE bifurcation analysis in Hughes et al., 2024.
The model accounts for active ($u$) and inactive ($v$) GTPase (such as Rac) and its nucleation of filamentous actin ($F$). We assumed that Rac promotes F-actin assembly, and that F-actin contributes negative feedback (inactivation of Rac).
The model equations are:
$$\begin{align}
\frac{\partial u}{\partial t} &= (b+\gamma u^2)v - (1+sF+u^2)u + D \Delta u \\
\frac{\partial v}{\partial t} &= -(b+\gamma u^2)v + (1+sF+u^2)u + \Delta v \\
\frac{\partial F}{\partial t} &= \epsilon (p_0+p_1 u - F) + D_F \Delta F
\end{align}$$
The dynamics of these equations in a 1D periodic domain, and XML files for simulating the model are given on model page M2071.
Here we investigate the effect of ‘actin wave’ dynamics on the protrusion and motility of a eukaryotic cell.
We assumed that the reaction-diffusion model operates on the 1D periodic cell-edge domain of a CPM cell. Using the Morpheus plugin StarConvex membrane based on the level of F-actin ($F$), we simulated the PDEs as a MembraneProperty system for a single cell, with parameter values given in the table below.
Polar Cell
Turning Cell
Parameters
$b$
$0.067$
$0.00067$
$\gamma$
$3.55$
$3$
$s$
$0.41$
$1$
$\epsilon$
$0.6$
$0.6$
$p_0$
$0.8$
$0.8$
$p_1$
$3.8$
$3.8$
$D$
$0.1$
$0.1$
$D_F$
$0.001$
$0.001$
lengthscale
$\frac{1}{5}$
$\frac{1}{100}$
Model file
Polar_Cell_ Motility_main.xml
Turning_Cell_ Motility.xml
Table 1: Parameter values for the Polar Cell and Turning Cell configurations. For the two-lobe (Two_Lobe_Cell.xml) and three-lobe (Three_Lobe_Cell.xml) configurations, the lengthscale parameter is set as described below, and all other parameters are as for the Turning Cell configuration.
To observe distinct behaviors, we picked cells of ‘different sizes’: we used similar kinetic parameters, but changed the lengthscale.
The lengthscale was obtained by adjusting the diffusion coefficients: in the Polar Cell, this coefficient was set to $\frac{1}{5}$; in the Turning Cell, to $\frac{1}{100}$; in the two-lobe configuration, to $\frac{1}{50}$ (twice the Turning Cell coefficient); and in the three-lobe configuration, to $\frac{1}{33}$ (three times the Turning Cell coefficient).
Results
Cell shape and motility vary depending on the type of PDE solution, determined by the parameter values and domain size, $L$. Four qualitatively different motility states are compared in Figure 1 (a)-(d) and the associated Videos 1-4 below. For each case, the corresponding model file is linked in the video caption.