Co-culture of cancer cells (darker cells) with stromal fibroblasts (lighter cells) with a signalling molecule shown in green speckles.

Introduction

In the last few years there has been an increasing awareness of the role matrix geometry plays in determining how cells move and cytoskeletal factors required for migration.

However, predicting migratory behaviour and the effect of experimental perturbation has proved difficult.

Inverted colour fluorescence image of a breast cancer cell.

Figure 1: Inverted colour fluorescence image of a breast cancer cell (MDAMB231) moving on a flat substrate. F-actin is shown in magenta, myosin IIa is shown in blue, and ezrin in yellow.

In the last few years there has been an increasing awareness of the role matrix geometry plays in determining how cells move and cytoskeletal factors required for migration. However, predicting migratory behaviour and the effect of experimental perturbation has proved difficult. Indeed, experiments have run ahead of our theoretical understanding of cell migration. To address this issue we collaborated with Melda Tozluoglu in the Biomolecular Modelling Group.

Melda developed a computational model that encompasses all key features of migrating cells and changing environments. Using the model, Melda first set up to identify the intracellular states that will provide fastest cell migration in different matrix geometries. This predicted the expected F-actin protrusion based migration strategies on unconfined surfaces, with the theoretical explanation for the roles of actin protrusions in maintaining contact with the surface and pulling the cell forward. Confined matrix geometries on the other hand, caused profound shifts in the relationship of adhesion and contractility to cell velocity.

The model predicted adhesion is dispensable in confined-discontinuous matrices and bleb driven migration will be an effective strategy. This provided explanation to the long lasting observation of adhesion-independent migration. We then tested our model with a highly challenging experimental context; the invasion of cancer cells in vivo.

Melda used the model to predict how cancer cells migrate into the discontinuous collagen matrix that surrounds tumours and the effect of different combinations of kinase inhibitors and integrin depletion in vivo. We used intravital imaging to test model predictions on bleb-driven migration, and predicted response to biochemical manipulations.

Strikingly, we could confirm the widespread use of bleb-driven cell migration in vivo and predict the effect of numerous experimental perturbations.

In addition to these studies, we are actively pursuing analysis of collective invasion of lobular breast cancers, the role of direct cell-cell contact between cancer cells and CAFs, intravital imaging of CAF signalling and the interplay between CAFs and chemotherapy. We hope that these studies will yield exciting insights into cancer biology and ultimately inform improved strategies for cancer therapy.