Date of Award

Summer 8-15-2016

Author's School

Graduate School of Arts and Sciences

Author's Department


Degree Name

Doctor of Philosophy (PhD)

Degree Type



In this thesis, we study the switch and pulse functions of actin during two important cellular processes, cell migration and endocytosis. Actin is an abundant protein that can polymerize to form a dendritic network. The actin network can exert force to push or bend the cell membrane. During cell migration, the actin network behaves like a switch, assembling mostly at one end or at the other end. The end with the majority of the actin network is the leading edge, following which the cell can persistently move in the same direction. The other end, with the minority of the actin network, is the trailing edge, which is dragged by the cell as it moves forward. When subjected to large fluctuations or external stimuli, the leading edge and the trailing edge can interchange and change the direction of motion, like a motion switch. Our model of the actin network in a cell reveals that mechanical force is crucial for forming the motion switch. We find a transition from single state symmetric behavior to switch behavior, when tuning parameters such as the force. The model is studied by both stochastic simulations, and a set of rate equations that are consistent with the simulations. Endocytosis is a process by which cells engulf extracellular substances and recycle the cell membrane. In yeast cells, the actin network is transiently needed to overcome the pressure difference across the cell membrane caused by turgor pressure. The actin network behaves like a pulse, which assembles and then disassembles within about 30 seconds. Using a stochastic model, we reproduce the pulse behaviors of the actin network and one of its regulatory proteins, Las17. The model matches green fluorescence protein (GFP) experiments for wild-type cells. The model also predicts some phenotypes that modify or diminish the pulse behavior. The phenotypes are verified with both experiments performed at Washington University and with other groups' experiments. We find that several feedback mechanisms are critical for the pulse behavior of the actin network, including the autocatalytic assembly of F-actin, the negative feedback of F-actin on Las17, and the autocatalytic self-assembly of Las17. These feedback mechanisms are also studied by a simple ordinary differential equation (ODE) model. Finally, we develop a partial differential equation (PDE) model that is more realistic than the ODE model and more computationally efficient than the stochastic model. We use the PDE model to explore the rich spectrum of behaviors of the actin network beyond pulses, such as oscillations and permanent patches. The predictions of the PDE model are of high interest for suggesting future experiments that can test the model.


English (en)

Chair and Committee

Anders E. Carlsson

Committee Members

Philip V. Bayly, John A. Cooper, Ralf Wessel, Li Yang,


Permanent URL: