Molecular mechanism responsible for transition of non-invasive breast cancer into invasive disease discovered

Protein allows cells altered by the oncoprotein ErbB2/HER2 to escape their bonds

Active, but non-invasive breast cancer is set free to roam as invasive breast cancer when an overexpressed protein converts it to a different cell type, scientists at The University of Texas M. D. Anderson Cancer Center report in the Sept. 9 issue of the journal Cancer Cell.

"We have discovered a key molecular mechanism for the deadly transition of non-invasive breast cancer into invasive disease," said senior author Dihua Yu, M.D., Ph.D., professor in M. D. Anderson's Department of Molecular and Cellular Oncology.

Overexpression of the protein 14-3-3ζ (zeta) launches a molecular cascade that removes bonds that tie the premalignant cells together and hold them in place, converting them from stationary epithelial cells to highly mobile mesenchymal-like cells, Yu and colleagues report. This epithelial-to-mesenchymal transition (EMT) is recognized as a crucial step in metastasis, the spread of cancer to distant organs that causes 90 percent of all cancer deaths.

The researchers show that 14-3-3ζ teams with the oncoprotein ErbB2, also known as HER2, in a two-hit process to convert normal mammary cells to invasive cancer cells.

In addition to identifying this key step in EMT, Yu notes the findings also provide:

• A biomarker in 14-3-3ζ to identify high-risk patients for more aggressive treatment before their noninvasive breast cancer converts to invasive disease.

• New therapeutic targets among the components of the molecular pathway launched by 14-3-3ζ. Some drugs already aim at these targets, Yu said.

• A solution to a puzzling mystery about how a subset of non-invasive breast cancer with excessive presence of a ErbB2/HER2 develops into invasive breast cancer.

Yu and colleagues previously showed that 14-3-3ζ is overexpressed in many other cancer types, like lung, liver, uterine, stomach cancers. "Our findings might have broader implications relating to the mechanism of invasion and metastasis in other types of cancer," Yu said.

Unzipping cancer cells

The team set out to address a longstanding puzzle, Yu said. ErbB2, an oncoprotein that promotes metastasis, is overexpressed in 50 to 60 percent of the noninvasive breast cancer known as ductal carcinoma in situ (DCIS). However, that same protein is overexpressed in only about 25 percent of invasive breast cancers, which seemed counterintuitive.

In a series of lab experiments, Yu and colleagues showed that overexpression of ErbB2 accompanied by overexpression of 14-3-3ζ can change DCIS into invasive breast cancer. This only occurs in about half of ErbB2-overexpressing DCIS, the team found, explaining the numerical puzzle.

Overexpression of ErbB2 converts normal breast duct cells into abnormal cells that reproduce quickly, are capable of moving, and resist programmed cell death that usually kills aberrant cells. What prevents these DCIS cells from becoming invasive, Yu said, is that they are locked together in zipper-like fashion by the cell surface protein E-cadherin, a trait known as cell-cell adhesion.

"Overexpression of 14-3-3ζ is the catalyst for a molecular pathway that strips E-cadherin from the cells, setting the cells loose from each other," Yu said. These cells also change in appearance from blunt normal breast duct cells to a narrow spindle shape characteristic of a mesenchymal-like cell.

Double overexpression reduces survival time

Epithelial cells line an organ or its cavities and are generally immobile. Mesenchymal cells are mobile and can differentiate into many different cell types, for example, to repair injury. Epithelial-mesenchymal transition is known to repress E-cadherin, decrease cell-cell adhesion and increase a cell's capacity to move. An estimated 80 percent of all solid tumors are carcinomas, cancers of the epithelial tissue.

Mice injected with a breast cancer cell line with both proteins overexpressed had three times the metastasis as mice with a control cancer cell line.

The researchers examined 107 invasive breast cancer cases and found that 23 of the cancers overexpressed both proteins. Those patients also had significantly shorter survival times due to metastasis-related deaths than those whose tumors expressed one or neither of the proteins.

Overexpressed 14-3-3ζ, the team showed, interacts with and stabilizes the receptor protein TβR1, which activates smad2/3 and moves them into the cell nucleus, where they in turn increase expression of ZFHX1B, which then represses expression of the adhesion protein E-cadherin.

Yu said that it will be very challenging to target 14-3-3ζ by drugs because it also regulates other important proteins in normal cellular processes. The downstream components such as TβR1 can be targeted with drugs that are under clinical trials.