Study sheds light on function of protein associated with high-risk breast cancers

The study identified how the protein, CARM1, recognizes its target proteins, and should lead to a better understanding of how the cancer develops as well as the discovery of targeted drug therapies.

“When CARM1 is overexpressed in breast cancer, the higher expression is correlated with a poorer prognosis,” said Dr. Wei Xu, professor of oncology at UWCCC and the McArdle Laboratory for Cancer Research. “But we know very little about how it works.”

CARM1 is a member of a related group of proteins known as PRMTs which all work by adding a small chemical modification, a methylation, to their target proteins, thus changing the function of the target. Previous work from Xu’s group has found a few cellular targets of CARM1, including one protein that promotes breast cancer metastasis when the protein has been chemically modified.

To understand how elevated CARM1 levels contribute to breast cancer growth, Xu wanted to know all its targets, not just a few. Then, she attended a UWCCC research meeting on campus, where both she and UWCCC member Dr. Joshua Coon, professor of biomolecular chemistry and chemistry, presented. Coon is also the director of the NIH National Center for Quantitative Biology of Complex Systems, a center housed at UW-Madison that specializes in protein identification techniques, including one known as quantitative mass spectrometry.

“After Wei presented, it became pretty clear that she had a problem that could benefit from our technology,” Coon said. “In the past, she could identify CARM1 targets, but it was one at a time. We can tell you all the proteins that are present in a sample, or, in her case, all the target proteins that are methylated by CARM1.”

Evgenia Shishkova, a graduate student who works with Coon through UW’s Integrated Program in Biochemistry, took on the project to identify all CARM1 targets in a highly efficient way. Rather than start with thousands of proteins and confirm or eliminate them one by one, she performed just two sets of experiments that isolated many new CARM1 targets.

Using breast cancer cells grown in the lab, Shishkova first isolated all proteins, then used molecular “scissors” to chop those proteins into smaller fragments. Next, she enriched that sample for all the protein fragments that had methylation modifications, and identified all those fragments using quantitative mass spectrometry. This set of experiments yielded hundreds of methylated targets – though not all of those proteins were methylated by CARM1.

“The second part was to repeat the same experiments, only this time in breast cancer cells that were genetically modified so that they did not produce any CARM1 protein,” Shishkova said. “By comparing these two conditions, we can say that whichever proteins are now absent when CARM1 is gone is going to be a target of CARM1. And of course then we had to prove they were indeed targets by other methods.”

Shishkova verified around 130 previously-unknown targets of CARM1. And because her experiments identified small fragments of those proteins, she was able to determine the exact short protein sequence CARM1 needs to “see” on its target proteins to modify them. The study also identified the part of CARM1 that recognizes this very specific target sequence.

“Many pharmaceutical companies are interested in developing inhibitors of CARM1, but they have not been successful because they focused on the part of the protein that is highly conserved among many PRMTs,” Xu said. “We hypothesize that we should focus on this unique region not found in the other PRMTs to develop inhibitors.”

Xu added that she and her group are already following up with many of these new targets to try to understand how modification of each target contributes to breast cancer development.

The study was published online May 24 in the journal Nature Communications. It was funded by National Institutes of Health grants P41 GM108538, R21 CA196653 and R01 CA213293, and by the National Cancer Institute’s core support grant to UWCCC, P30CA014520-UW.