WashU biomedical engineer combines data, algorithms to understand HER2 breast cancer gene

In American women, breast cancer is the second most common cancer and the second leading cause of cancer death. Using data, algorithms and lab experimentation, a biomedical engineer at Washington University in St. Louis is studying breast cancer at the most basic level – the cells – to look for clues about how the cancerous cells metastasize.

Kristen Naegle's research combines computational mining and modeling techniques with experimental molecular biology approaches to understand the function of post-translational modifications in regulatory networks of the cell.

Kristen Naegle, assistant professor of biomedical engineering in the School of Engineering & Applied Science, applied her unique computational skills to look at the HER2 gene. HER2-positive breast cancers are aggressive and spread faster than other types. Researchers have found that too much protein is made from the HER2 gene — called overexpression — in 20 percent of all breast cancers, making HER2 a valuable target for potential personalized treatment methods for this type of breast cancer.

To determine why HER2-positive cancers are more aggressive, Naegle analyzed measurements from a previous study that isolated signaling molecules in a HER2-overexpressing breast cell line and a normal breast cell.

“We use mathematical approaches to find similarities in the data,” Naegle said. “For this dataset, we looked at how signaling molecules are most related to each other in the normal cells, compared to how they are related to each other in the HER2-overexpressing cells. We looked for relationships that are drastically different in the two cell types to understand how signaling is altered. Despite the fact that individual molecules are highly similar to each other across cell types, we found that small changes in signaling dynamics led to very large changes in the relationships uncovered between groups of signaling molecules.”

One of these big changes they found involves a protein that regulates how cells are connected.

“One of the things that decreases metastatic behavior is that the cells stick tightly together through cell-cell junctions,” Naegle said. “That told us that if there are signaling alterations happening at the cell junctions, then maybe that’s why these cells are more metastatic.”

Naegle and members of her lab tested the hypothesis that interactions with this cell junction protein were altered according to the differences in the signaling relationships they saw from their analysis. They found that interactions with the cell junction protein were very different between the two cell types, and the interaction dynamics matched the dynamics of the signaling that uncovered the relationship.

“This is exciting, because it’s been proposed that testing for interactions in a cancer biopsy may be a better predictor of how a cancer will respond to a treatment,” Naegle said. “Given that some HER2-positive breast cancer patients don’t respond to HER2 therapy, maybe this protein interaction could help us identify patients who will respond well to therapy and those who will not gain additional benefits.”

Additionally, with Venktesh Shirure, a former research scientists in biomedical engineering, the team conducted a series of experiments to test whether cell junctions were altered in the conditions that correspond with the most aggressive cell type. Their experiments revealed that the cell junctions in with high HER2 expression break down and become “leaky” in response to the growth factor, which may be related to how HER2-postive cancers metastasize.

“This study shows that by using a different mathematical interpretation of the data, even a decade after its first publication, we can still find new nuggets of information hidden in these high-throughput, systems-level measurements of early signaling dynamics and identify novel, unknown findings,” Naegle said.

“My lab believes that disease is a context shift, so what we should fundamentally understand is how context shapes cell decisions and then understanding disease becomes relatively trivial,” Naegle said. “It’s a bottom-up approach where we look to understand the basic mechanisms of the interactions in the cell to find the outcomes. There still remains a wealth of hypotheses from this analysis that may continue to help us understand how HER2 drives metastasis.”


Schaberg K, Shirure V, Worley E, George S, Naegle K. “Ensemble clustering of phosphoproteomic data identifies differences in protein interactions and cell-cell junction integrity of HER2-overexpressing cells.” Integrative Biology, April 28, 2017. 2017, 9, 539. DOI: 10.1029/c7ib00054e.


The McKelvey School of Engineering at Washington University in St. Louis promotes independent inquiry and education with an emphasis on scientific excellence, innovation and collaboration without boundaries. McKelvey Engineering has top-ranked research and graduate programs across departments, particularly in biomedical engineering, environmental engineering and computing, and has one of the most selective undergraduate programs in the country. With 140 full-time faculty, 1,387 undergraduate students, 1,448 graduate students and 21,000 living alumni, we are working to solve some of society's greatest challenges; to prepare students to become leaders and innovate throughout their careers; and to be a catalyst of economic development for the St. Louis region and beyond.