Triple negative breast cancer cells use new strategy to boost metastatic ability

Researchers at Baylor College of Medicine have uncovered a strategy that triple negative breast cancer (TNBC) cells use to boost their ability to metastasize, or spread to other organs. Metastasis is the leading cause of cancer-related deaths and scientists are investigating ways to prevent it. These findings, published in Nature Communications, highlight new possibilities for developing clinical interventions to treat metastatic TNBC patients for whom there are no specific therapies.

"Metastasis occurs when cells break away from the main tumor and travel through the bloodstream to spread to other parts of the body where they can seed new growths," said corresponding author Dr. Chonghui Cheng, professor of molecular and human genetics and molecular and cellular biology at Baylor. Cheng also is a member of the Lester and Sue Smith Breast Center and the Dan L Duncan Comprehensive Cancer Center, both at Baylor. "Scientists know that circulating tumor cells (CTC) are more likely to give rise to new tumors when they travel in clusters rather than as single cells. Clusters survive better and lodge in new organs more easily."

But there is a mystery surrounding TNBC clusters. Typically, adherens junction (AJ) proteins mediate stable cell-cell interactions that keep clusters together. But AJs often are lost in highly aggressive TNBCs, raising the question of how CTCs form clusters from such tumors.

Cheng and her colleagues compared TNBC cells with non-TNBC cells as well as metastatic and non-metastatic breast cancers. "When we analyzed the data, the extracellular matrix stood out, particularly hyaluronan (HA), one of its components," said co-first author Dr. Georg Bobkov, instructor in the Cheng lab.

The extracellular matrix is like a 'sticky coat' around cells. It's a complex structure made of proteins, carbohydrates and water that acts like scaffolding and 'glue' that holds cells together.

Focusing on mouse models of tumor metastasis and patient blood, the researchers isolated CTC clusters and found that TNBC cells produce large amounts of HA and coat themselves with it. "This HA coat is produced by an enzyme called HAS2, which TNBC cells make in unusually high amounts," said co-first author Dr. Khushali Patel, postdoc in the Cheng lab. "When we removed HA from CTCs, the clusters did not hold together. We also found that HA works with a cell surface protein called CD44. Without CD44, HA cannot be presented on the cell surface and cells cannot cluster."

After HA and CD44 connect on the cell surface, their interaction is further stabilized by another set of proteins called desmosomes. This allows the cells to form strong clusters that can sustain sheer stress with little damage as they travel in the bloodstream. Cancer cells are protected when they are clustered, which enhances their ability to propagate to other organs.

There is another benefit from the HA-based clustering. While AJ-mediated clustering is rigid, HA mediates the formation of more flexible clusters, which become an advantage in certain situations. We knew from other studies that when CTC clusters travel through very narrow blood vessels called capillaries, they do so in a single-cell file in which the cells stay in contact. They reform the cluster after they pass the capillaries. Our findings found an explanation for this behavior. The flexible nature of HA-mediated clustering allows CTC clusters to disassemble temporarily while the cells cross a narrow path and reassemble into the protective cluster afterward."

Dr. Georg Bobkov, co-first author 

HA-mediated clustering offers another advantage to CTCs. Researchers knew that CTC clusters can be a combination of cancer cells and other cells, including immune cells. "Some of the immune cells in clusters, like neutrophils, protect CTCs, but we did not know how clusters sequestered neutrophils," Patel said. "Neutrophils express CD44 on their surface. HA on clusters can bind to it, capturing neutrophils."

Although more research is needed, Cheng and her colleagues believe that their findings open new possibilities for treatments. "One idea is to develop ways to prevent the formation of or to disassemble CTC clusters by blocking the binding between HA and CD44, which we expect would prevent or reduce metastasis," Cheng said. "We also found HA-CD44-mediated clustering in glioblastoma, prostate and pancreatic cancers, so future therapies could potentially benefit patients with other metastatic cancers."

Other authors of this work include Bree M. Lege, Rong Zheng, Gad Shaulsky and Matthew J. Ellis, all at Baylor College of Medicine.

This research was supported by grants from NIH T32GM08307, T32ES027801-06, T32ES027801-05 and R01CA276432, Frank & Sandra Kimmel Postdoc Endowment and a Cancer Prevention Research Institute of Texas Scholarship in Cancer Research (RR160009). The following grants provided further support: NIH DK56338, CA125123, ES030285, S10OD030414, S10OD020151, CPRIT RP150578, RP170719, NIH P50HD103555, NIH 1S10OD016167, NIH CA125123, RR024574 and CPRIT RP180672.