BioWorld International Correspondent

LONDON - Skin cancer cells recruit other cells from the skin to make a track that will allow them to invade further into the body, a new study has shown.

The discovery creates a new paradigm of how cancer cells invade surrounding tissue, and could lead to different therapeutic approaches for preventing metastasis.

Erik Sahai, principal investigator at the Cancer Research UK London Research Institute, told BioWorld International: "We have found that cancer cells use non-cancer cells to dig holes into the surrounding tissue. The cancer cells then follow behind the cells that have done all the work, a bit like getting a piggyback ride."

Sahai and his colleagues reported their findings in a paper in the advance online edition of the Nov. 25, 2007, Nature Cell Biology, titled: "Fibroblast-led collective invasion of carcinoma cells with differing roles for RhoGTPases in leading and following cells".

Pathologists have known for decades that tumor cells from cancers arising in epithelial tissues, such as the skin, breast, liver or bowel, sometimes acquire characteristics of fibroblasts - a process called epithelial-to-mesenchymal transition. In experimental systems, tumor cells with fibroblast-like qualities were capable of invading underlying tissues and metastasizing to distant parts of the body. However, many invasive carcinomas do not "change their spots" in that way; instead, they retain many of their epithelial characteristics.

Sahai and his colleagues set out to make sense of those conflicting observations. They took cells from squamous cell carcinoma (SCC), a type of skin cancer, and grew these on top of a matrix containing different types of collagen. The top of the matrix was exposed to the air, to mimic skin.

When the researchers placed SCC cells on top of the matrix, they differentiated in an abnormal way, but stayed on top of the matrix. When the scientists added fibroblasts that had been isolated from oral or vulval SCCs, however, the SCC cells - which continued to express epithelial cell markers on their surfaces - invaded the underlying matrix.

The more fibroblasts that the researchers added, the more the SCC cells were able to invade the matrix.

Sahai said: "What we showed was that, instead of the SCC cells behaving like fibroblasts themselves, they recruited the fibroblasts to do all the hard work of remodeling the matrix by making a hole in it. We could observe chains of cells moving within the holes or tracks, each made by a fibroblast."

In each case, he added, the fibroblast was at the front, with a line of SCC cells following behind.

Further experiments conducted by the team made it possible to identify which enzymes the fibroblasts used to allow them to make the tracks; studies showed that those enzymes were different to those used by the SCC cells, which allowed the tumor cells to move along the paths created by the fibroblasts.

Sahai and his colleagues went on to investigate what would happen when they inhibited the enzymes used by the fibroblasts for remodeling the matrix. They found that blocking the action of integrin alpha3, integrin alpha5 or Rho-ROCK in the fibroblasts greatly reduced the ability of the cells to generate tracks in the matrix. As a result, the SCC cells could no longer follow behind the fibroblasts.

"Interestingly, some of the inhibitors we used to block the enzymes known as ROCK kinases had already been developed as potential anti-hypertensive agents," Sahai said.

If investigations were to show that such inhibitors could reduce the ability of cancer cells to invade and metastasize, however, it would be important for researchers to consider at what stage those drugs should be given, Sahai reflected. "If such drugs were given after fibroblasts have made the tracks in the surrounding tissue, then giving such a blocker would not be expected to make much difference. This may be one reason why other trials of anti-invasion agents have not met with much success," he said.

The study reported in Nature Cell Biology adds an "unwelcome complexity" to the question of how best to prevent tumor metastasis, Sahai said. "It shifts our focus," he added. "We need to consider which cells we should be studying and which systems, if we really want to stop invasion. It tells us that, if you look at cancer cells in isolation, you may miss the point."

Future work by the team will aim to find out if similar events occur to assist invasion by other types of cancer cells.