“Biologists and physicists who study motility have mainly focused on investigating how large collectives comprising hundreds or thousands of cells coordinate their movements,” said David Brückner, doctoral student in the lab of Chase Broedersz, PhD, associate professor of theoretical biophysics at Ludwig-Maximilians-Universitaet (LMU) in Munich and the Vrije Universiteit (VU) in Amsterdam.

“We wanted to know how pairs of cells interact when they come into contact and set out to analyze their behavior using the methods of statistical physics.”

All biological processes involve the movement of cells. And when cells move en masse, they are bound to encounter each other. During embryonic development, cells communicate with their neighbors to find their place in the growing organism. When wounds heal, cells pick up cues to move into lesions and regenerate lost or damaged structures. The spread of cancer requires cancer cells to find blood vessels that will transport them away from their origin.

Although much is known about the molecular basis of cellular interactions, system-level dynamics of the behavior of interacting cells are poorly understood.

Brückner and his colleagues have created a microscopic “cage,” they call a “cell collider.” The use of defined geometries is a popular strategy in studies of cell motility, but Alexandra Fink, a PhD student in the group led by Joachim Rädler, PhD, professor of biophysics and the physics of soft matter at LMU, has now applied it in a novel context.

“The idea was to isolate two cells, while permitting them to interact in a restricted fashion,” said Brückner.

The authors placed single cells in two compartments connected by a narrow channel in the cell collider: a dumbbell-shaped structure in which pairs of cells are made to collide repeatedly.

This geometry forces cells to interact by extending thin membrane projections through the narrow channel, inevitably leading to collisions. The cell nuclei are fluorescently tagged to enable the movements of both cells to be tracked by microscopy over time.

“Based on the resulting experimental data, we were able to develop a model that provides a physical description of how the cells interact,” said Brückner.

When normal cells come into contact in the cell collider, their membranous protrusions repel each other, and are then fully retracted.

“When normal cells make contact, they tend to change direction so as to avoid the obstacle,” said Brückner. This response to initial contact enables the cells to keep their distance from one another. “We were surprised to find that tumor cells behave in a very different way.”

When cancer cells meet, they almost always try to get past each other. With the aid of their theoretical model, the team simulated this response in great detail. The analysis revealed that when two tumor cells approach each other they do not slow down as normal cells do. Instead, they speed up and try to squeeze past one another.

Based on the trajectories of pairs of normal and cancerous cells in the confines of the cell collider, the authors infer an interacting equation of motion, which accurately predicts characteristic pairwise collision behaviors of different cell lines, including reversal, following, or sliding events.

The authors find interacting noncancerous MCF10A cells can be described by repulsion and friction interactions while interacting cancerous MDA-MB-231 cells can be described by attraction and anti-friction interactions, promoting the predominant sliding behavior observed in these cells.

“These findings suggest two interesting approaches for further investigations,” said Brückner.

In future studies, the LMU team plans to identify the molecular bases for the very different interactions of the two cell types. Cancer cells express a different set of surface proteins compared to their counterparts.

Among them, cadherins—a specific class of proteins that play a vital role in mediating cell adhesion—are of particular interest. Further studies will explore whether cadherins are sufficient to account for the differences observed in cancer and normal cell interaction dynamics.

In addition, the authors intend to examine whether bigger cell aggregates, such as those found in tumors, display patterns of motility like those exhibited by pairs of cells.

This approach can be used to quantify the effect of molecular disruptions on the interactions of pairs of cells or large cell clusters, leading to insights on cell migration.