Polarity protein suppresses tumor growth, invasionby Leigh MacMillan | Nov. 29, 2012, 8:59 AM
Ninety percent of human cancers arise from epithelial cells — cells that form surfaces in many tissues, including the lungs, gut and skin. Epithelial cells have a characteristic shape, with different “top” and “bottom” surfaces, and this “polarity” is usually lost during tumorigenesis. It has not been clear, however, if the machinery that builds the top-bottom polarity plays a role in cancer.
Now, studies led by Ian Macara, Ph.D., who recently joined the Vanderbilt faculty as chair of Cell and Developmental Biology, demonstrate that loss of the polarity protein Par3 promotes breast cancer formation and metastasis in a mouse model. The investigators also discovered that Par3 expression is reduced in human breast cancers. The findings, reported in the Nov. 13 issue of Cancer Cell, establish Par3 as a suppressor of tumor growth and invasion.
Par3 is one of a group of proteins that function as master regulators to produce top-bottom polarity in epithelial cells. Macara and his colleagues have focused for many years on the Par proteins and how they work.
“It is widely assumed that loss of cell polarity is a key step in cancer progression, but in fact there is very little evidence to support this assumption,” Macara said. “Because of our interest in the polarity machinery, we set out to ask whether it’s actually involved in cancer.”
The researchers used a novel mouse model of cancer: they isolated stem cells from the mammary gland of the mouse, infected the cells with a virus carrying specific genetic changes, and transplanted the cells back into the mouse mammary fat pad. The altered stem cells multiply and generate a mammary gland with all of its cell types.
In this model, introduction of an oncogene (a gene that promotes cancer) causes the formation of small breast tumors that don’t metastasize, Macara said.
But when the researchers also reduce expression of the polarity gene Par3, the mice develop large breast tumors and “huge numbers of metastases to the lungs,” he said.
“It was quite a dramatic difference.”
Macara and his team characterized the molecular consequences of Par3 depletion, and discovered increased activity of signaling pathways that stimulate invasion by causing degradation of the extracellular matrix that surrounds cells.
“These polarity proteins are providing spatial information. Inside the cell, it’s like a city — things have to be in certain places, and if they get moved to the wrong place at the wrong time it can cause a lot of trouble,” Macara explained.
In studies of human breast tumors, the researchers found that more than 50 percent of the samples had lost expression of the Par3 gene, and the human tumors lacking Par3 had the same inappropriate activation of invasive signaling pathways.
International cancer genome sequencing efforts have also found mutations in the Par3 gene, supporting its importance in the formation of human cancers.
The mis-activated pathways triggered by loss of Par3 may contain potential therapeutic targets for blocking tumor growth and metastasis, Macara said.
“We think that the mouse model is providing us with a very good model of what’s happening in human breast cancer,” he said. “So finding drugs that inhibit these pathways will be an important future goal.”
One big surprise of the current studies was that the metastatic cells lacking Par3 continue to express molecular markers of the epithelial cell “type” and move together in small clusters rather than as single cells. This finding contrasts with the classical view that during tumorigenesis epithelial cells lose the ability to stick to one another and change into a different type of cell, which moves as an individual cell rather than as part of a group.
“Our mouse model has generated some interesting new ideas about how cancer cells can spread through the body,” Macara said.
Macara is the Louise B. McGavock Professor. Authors of the Cancer Cell paper include Luke McCaffrey, Ph.D., JoAnne Montalbano, Ph.D., and Constantina Mihai.
The research was supported by grants from the National Institutes of Health (GM070902, CA132898, CA139950) and from the Terry Fox Research Institute.
Leigh MacMillan, (615) 322-4747