Scientists have found a new clue to an important question in cancer research: how do cancer cells spread? The clue lies with changes in their stickiness or adhesion properties: they become unstuck at the original tumor site, then reattach themselves at a new site. The changes involve molecular interactions between cells and the extracellular matrix, the “scaffolding” that holds cells in place to form three-dimensional tissue.

Such discoveries are important because 90% of cancer deaths are caused not by primary tumors but by metastatic tumors, the ones that grow from cells that have travelled from the original site to another vital part of the body.

Writing in the 9 October online issue of Nature Communications, study leader Sangeeta Bhatia and colleagues, say their findings offer potential new targets for cancer drugs.

Bhatia, a biological engineer and professor at the Massachusetts Institute of Technology in the US, says in a press statement:

“As cancer cells become more metastatic, there can be a loss of adhesion to normal tissue structures. Then, as they become more aggressive, they gain the ability to stick to, and grow on, molecules that are not normally found in healthy tissues but are found in sites of tumor metastases.”

Finding a way to stop them growing at the new sites could interfere with metastatic disease, she adds. This could be enough to impede the growth of secondary tumors.

The scaffolding that helps cells form a three-dimensional structure in tissue is the extracellular matrix. The matrix also plays an important role in controlling what cells do. Proteins called integrins sit on the surface of cells and behave like anchors, they keep the cells tethered to the matrix.

But when cancer cells get ready to migrate (metastasize) the integrin anchors let go. In their study, Bhatia and colleagues found out how.

They genetically engineered mice to develop lung cancer, then took cells from four sites in their bodies: from primary lung tumors that later spread (metastasized), from primary lung tumors that did not metastasize, from metastatic tumors that migrated from the lungs to nearby lymph notes, and from metastatic tumors that travelled to sites further away, like the liver.

Using a method they had developed a while ago, which much improved their ability to study cell adhesion, they exposed each type of cell to around 800 pairs of molecules from the extracellular matrix.

They put tiny spots of the cells onto microscope slides and to each added two different proteins from the extracellular matrix. They could then examine how well the cells from each type of tumor stuck (adhered) to the protein pairs.

The new method gave them the ability to study interactions of many more molecules with more cells than before. Also, by allowing them to study pairs of molecules, they could look for adhesion synergies, where molecule pairs work together to anchor the cell.

A real surprise came when the team discovered that where adhesion properties were concerned, metastatic cells from different primary tumors were more similar to one another than to the other cells in the primary tumor they came from.

One pair that stood out as being particularly “sticky” for metastatic tumors was fibronectin and galectin-3. Both proteins contain or bind to sugars.

Lead author Nathan Reticker-Flynn, a PhD student in Bhatia’s lab, says although metastatic tumor cells share adhesion traits, they may take different pathways to get there.

Some tumor cells appear to change the combination of integrins they express, while others vary the types of sugars on their surfaces.

The result is that all these changes lead to higher or lower stickiness (affinity) for certain molecules in the extracellular matrix.

The researchers found similar results when they examined primary and metastatic human tumor samples. One thing they noted was that the more aggressive the metastasis, the more galectin-3 was present.

There is already evidence that tumors prepare a remote site for migration by sending out molecules that change the environments of those sites to make it easier for cancer cells to grow there. The researchers suggest part of this process could be accumulation of galectin-3 and other molecules.

“There’s a lot of evidence to suggest that a hospitable niche for the tumor cells is being established prior to the cells even arriving and establishing a home there,” says Reticker-Flynn.

The researchers suggest their findings offer potential new targets for blocking metastasis. An alternative to focusing on a particular gene mutation, would be to focus on a particular protein-protein or protein-sugar interaction.

“If those changes do confer a lot of metastatic potential, we can start thinking about how you target that interaction specifically,” says Reticker-Flynn.

The team has already had a go to see if this works: they knocked out a gene so less integrin was expressed on the surface of cancer cells that they had identified as interacting with the fibronectin and galectin-3 pair. The tumors did not spread as much in those mice.

Another possibility for drug development is to block the binding sites on fibronectin and galectin-3 with antibodies, so tumor cells can’t anchor to them.

The team is now trying to discover more details about how tumor cells interact with galectin-3, and is developing a list of candidate drugs that might be able to stop those interactions.

Funds for the study came from Stand Up to Cancer, the Koch Institute Circulating Tumor Cell Project, the Harvard Stem Cell Institute, the National Cancer Institute, the Howard Hughes Medical Institute and the Ludwig Center at MIT.

Written by Catharine Paddock PhD