Utilization of mechano‐biological models to predict cell adhesion interactions between bone marrow endothelial cells and breast cancer cells.

The purpose of our study was to produce a mechano‐biological model that could characterize and predict adhesive interactions between human breast cancer cells (MB435) and human bone marrow endothelial cells (HBME). Bone is a common site of metastasis for breast cancer and the mechanisms of metastasis are not fully characterized. In this study we have combined experimental and modelling approaches to probe interactions of human bone marrow endothelial cells (HBMEC‐60) with breast cancer cells (MB435). Specifically, atomic force microscopy (AFM) was used to measure adhesion forces between a single MB435 cell (attached to the AFM tip) and a monolayer of HBMEC‐60 cells for differing times of contact. In addition, we developed a stochastic model of the individual adhesive interactions that occur between the two cell‐types to mimic the AFM experimental protocol. This was accomplished by formulating a 2D discrete stochastic‐elastic computational model to study the adhesion binding dynamics that underpin cell‐cell attachment. The model allows detailed quantification in binding dynamics and the location, lifetime, and strength of cell‐cell adhesions. To describe cell‐cell interactions, we considered the following components within a discrete stochastic computational framework: intracellular actin in each cell and adhesion proteins in the cell membranes in 4 possible states: freely diffusing, actin‐bound to the first cell, actin‐bound to the second cell and fully‐bound to cytoskeleton within both cells. Adhesions between the cytoskeleton networks of the two cells were connected to form a network of springs according to the transitions, which obey stochastic binding/unbinding events. Free adhesion molecules diffused via a random walk. The dynamically changing network structure was coupled to the local mechanical environment through force‐dependent binding/unbinding propensities. Experimentally, we found that peak adhesion forces increased with increasing contact time between the MB435 and HBMEC‐60 cells and that total adhesion force increased dramatically within the first 30 sec of contact which then gradually plateaued. These experimental data demonstrate a strong dependence of MB435‐HBME adhesion strength on cell‐cell contact time. Additionally, we found that our stochastic computational model recapitulated these experimental findings, suggesting that important mechanisms have been captured in the model. This validates our experiment vs. model approach and is therefore allowing us to interrogate the model further to derive further insights into the cell‐cell interactions observed here.