How mechanical forces between cells influence where cancer develops

Cancer initiation and spread are governed as much by mechanical context as by genetics. The physics of the tissue plays an important role in both triggering tumour development in one tissue and not in another, and in determining whether cancer progresses

Why do some tissues develop a tumour while others remain unaffected by the same mutation? In our recent study published in eLife, we discovered that the answer may not lie in genes but in physics, in the subtle mechanical forces that govern how our cells stick, pull, and push against one another. The physics of biological tissues plays an important role in both cancer initiation, and in determining whether the cancer progresses.  

Our findings show that differences in interfacial tension — the physical force acting along the boundary of healthy and cancerous cell populations — explains why a cancer-causing gene triggers tumour-like growth in human in one tissue but not in another. These differences in cellular mechanics provide a physical basis for tissue-specific oncogenesis, offering new insights into how the earliest stages of cancer may be regulated by the mechanics of the tissue itself.

We have looked at cancer initiation as this is still under explored in most cancer studies; a lot of studies focus on tumour growth and metastasis. We have looked at the precancerous state — before the tumour has even formed. The study shows that the properties of the tissue are just as important in determining whether cancer will initiate.

Where mutations meet mechanics

More than 80% of human cancers begin in epithelia, the thin sheets of cells that line and protect our organs. Despite their similar roles, each epithelial tissue has distinct architecture and mechanical properties. Could inherent mechanical differences influence how tissues respond to oncogenic mutations? In other words, whether the physics of a tissue could make it more or less susceptible to cancer.

The mutation studied, HRas^V12, is a permanently “on” version of the normal Ras gene that constantly signals affected cells to proliferate and survive. Mutations in Ras genes are among the most common in human cancers, including those of the lung, pancreas, and skin. However, despite its potent growth signal, HRas^V12 expression alone is not always sufficient to form tumours, suggesting that other tissue-specific factors, such as the local mechanical environment may influence its outcome.

Interestingly, while HRas^V12 mutations rarely drive breast cancer (found in less than 1% of breast ductal carcinomas), its close variant, KRas^V12, is highly prevalent in lung cancers.

To investigate why the same oncogenic signal leads to such different outcomes, we recreated early cancer-like conditions in human mammary (breast) and bronchial (lung) epithelial systems by introducing the HRas^V12 mutation in both systems and observed how the mutation behaved in each. The difference was striking. In mammary epithelium, HRas^V12 mutant cells formed small, compact clusters that stayed confined. But in bronchial epithelium, the same mutation produced spreading, invasive clusters that overgrew their surroundings.

Tension at the borders

When a combination of microscopy and advanced image analysis tools were used, it became clear that in mammary epithelium, a dense “belt” of contractile actomyosin fibres — a network of two proteins, actin and myosin — formed around the mutant clusters. Actin forms long, flexible filaments that serve as cellular “tracks,” while myosin acts like molecular motors that grab and pull on these filaments, generating contractile forces within cells. This contractile actomyosin belt around the mutant clusters compressed them into tight “bubbles” and restrained them like a tightening drawstring and prevented their spread.

In the bronchial epithelium, however, cells were more elongated and loosely connected. Here, no contractile belt formed, allowing the oncogenic clusters to be not constrained or spread freely and invade neighbouring healthy regions, forming finger-like protrusions reminiscent of early tumour invasion.

The tightening of mutant “bubbles” in mammary tissue hinted at an underlying mechanical cause — especially since actomyosin networks are the cell’s built-in machinery for generating tension and shape.

To probe deeper, we teamed up with physicists at IISc and implemented a vertex model of epithelia, modifying it to capture our experimental observations. In this model, the tissue is represented as a mosaic of polygons (cells) connected by edges (cell junctions). The system’s total energy depends on cell shapes and the tension along their shared borders, and it continuously reorganises to minimise this energy. Our key tweak was to include two cell types — mutant and healthy — and introduce a non-zero interfacial tension between them, then vary its strength.

The simulations showed that when this interface carried higher tension, the mutant clusters started to tighten and compact, but a negative interfacial tension destabilised the boundary, allowing mutants to spread and invade. As per the equation of the vertex model, this would seem quite intuitive — a higher interfacial tension tightens the boundary, bringing down the overall energy of the system. But what excites us as biologists who borrow tools from physics to tackle fundamental biological problems is how such a simple physical principle can explain something as complex as the earliest steps of cancer initiation.

This mathematical framework provided a simple physical explanation for the observations that the interfacial tension between oncogenic and healthy populations alone can determine whether mutant cells remained confined or became invasive.

Finally, to confirm this experimentally, the contractile belt in mammary epithelium was disrupted using blebbistatin, a drug that relaxes the binding between actin and myosin. Once the belt was weakened, the previously confined clusters expanded and began behaving like those in bronchial tissue. This confirmed that mechanical tension, driven by the actomyosin belt around the mutant clusters, is what restrains them, preventing invasion in mammary tissue but not in the bronchial epithelium.

Therefore, the nature of mechanical interactions between healthy and oncogenic counterparts can determine the outcome of the mutation.

When physics guides cell fate

These results highlight how physics and biology intertwine in the earliest stages of cancer. Oncogenic mutations not only alter biochemical pathways but also reshape the physical landscape of the tissue.

Whether a precancerous mutation turns dangerous largely depends on how the neighbouring healthy cells mechanically respond — whether they are able to push back or succumb.

A new angle on cancer prevention

Our findings suggest that cancer initiation is governed as much by mechanical context as by genetics. Healthy epithelial tissues possess a natural ability called epithelial defense against cancer, in which normal cells recognise and eliminate abnormal ones.

However, this defense weakens when the tissue’s mechanical balance is disturbed as seen during chronic inflammation, smoking, or prolonged pollution exposure. Strengthening or preserving this balance might therefore prevent cancerous mutations from taking root.

Understanding how mechanical tension acts as a barrier could open new paths to early intervention — perhaps even allowing us to reinforce these protective forces before tumours form.

Looking ahead

The next step is to explore how other oncogenes behave across tissues with different mechanical environments, and to identify the molecular players that tune interfacial tension, from actomyosin networks to cell adhesion molecules.

By decoding how cells use physical forces to maintain tissue harmony, we may learn how to predict where cancers are likely to start and how to stop them before they begin.