Profile of Myles Brown

Ever since oncologist Myles Brown used tamoxifen to treat a patient with metastatic breast cancer during his fellowship, he has wanted to understand how cancer cells can sometimes become resistant to this therapy after months of success. That quest led him to his life’s work on estrogen receptor (ER) biology. Brown has methodically pieced together the complex dance between steroid hormones such as estrogen, the receptors that bind them, and DNA. In the process, he has discovered molecules that enhance RNA transcription and uncovered the steps that lead to transcription. His goal is to contribute to novel and effective cancer therapies. His Inaugural Article (1) harnesses the power of CRISPR, a gene editing tool, not only to confirm many of his findings from the past 20 years but also to enhance understanding of the complex feedback loops between ERs, transcriptional enhancers, and genes that play a role in estrogen-positive breast cancers. Brown, the Emil Frei III Professor of Medicine at Harvard Medical School and Director of the Center for Functional Cancer Epigenetics at the Dana–Farber Cancer Institute, was elected to the National Academy of Sciences in 2016.

Myles Brown. Image courtesy of the Dana–Farber Cancer Institute.

Finding Direction

Brown grew up in Bethesda, Maryland near the NIH. He lived his first five years in Massachusetts with his mother, his two brothers, and his father, who was a gynecologist. After his father died from Hodgkin’s lymphoma, his mother remarried British neurosurgeon Sydney Green and the family moved to Bethesda. The move shaped Brown’s life. In the fall of 1973, during his senior year at Walt Whitman High School, a friend who worked at the NIH told him that researcher George Khoury was looking for help. Brown had gravitated toward biology, but it was his decision to work for Khoury that provided much-needed direction.

“At the time, George was a junior investigator at NIH,” recalls Brown. “He was part of a generation of physician-scientists who ended up at NIH during the Vietnam War . . . He was my first and most important mentor.”

Brown decided to emulate Khoury’s career path, using medical school as a launching pad toward science. He worked with Khoury through his senior year and summers, while he completed his bachelor’s degree in biology at Yale University. Khoury’s research focused on simian virus 40 (SV40), a contaminant of the polio vaccine found to cause cancer in animals. Khoury’s laboratory was mapping SV40 transcription patterns and was among the first to discover transcriptional enhancers, says Brown.

“I was at NIH during the height of the molecular biology revolution,” recalls Brown. “It was extremely exciting. Even as a high-school student, I could do things that actually ended up with authorship on papers” (2, 3).

Brown was not the best student at Yale University, he admits. However, his experience working in Khoury’s laboratory helped secure him a spot at Johns Hopkins University School of Medicine. His goal was to become a disease-oriented laboratory researcher.

“I do laboratory research with the overarching goal of solving questions that are relevant to breast and prostate cancer specifically, but also cancer in general,” says Brown. “The advantage of having had medical training is that I have an understanding of clinical problems and clinical implications of what we do, and that helps inform how we focus our work and why certain questions are important to pursue.”

Finding a Focus

Attending medical school in Baltimore allowed Brown to continue working with Khoury until he graduated in 1982. He then moved to Brigham and Women’s Hospital in Boston for a four-year research residency that included an internal medicine residency and a research fellowship with his second mentor, David Livingston, at the Dana–Farber Cancer Institute. Livingston also worked on SV40 and its link to cancer. Brown’s work with Livingston on the transforming properties of the SV40 large T antigen and the development of a system to regulate genes in animal cells using the lac repressor and the SV40 promoter led to his first lead-author publications (4, 5).

After his residency, Brown took a fellowship in medical oncology that included an intensive clinical year at Dana–Farber Cancer Institute and three years as a postdoctoral fellow in molecular biologist Phillip Sharp’s laboratory at the Massachusetts Institute of Technology (MIT). By this time, Brown needed a research question of his own. An encounter with a patient with metastatic breast cancer provided inspiration. Brown treated her cancer with tamoxifen, and her tumors shrank, only to return months later. Brown wondered how cancer cells that depend on estrogen to divide become resistant to drugs that block estrogen. He decided to focus on the ER, a transcription factor that helps determine which genes a cell turns on.

“Any given cell’s identity is driven primarily by the expression of a relatively small set of transcription factors that together, in a network of reinforcing expression, determine the identity of that cell type,” explains Brown. ER is one such factor in breast cells.

Since much of Brown’s work on SV40 focused on transcriptional regulation, it was an easy transition to start working on the ER. In Sharp’s laboratory at MIT, he used tools pioneered there, including gel shifts, to study transcription factors. His first paper on ER identified the ability of ER to form distinct protein–DNA complexes depending on whether it was bound by estrogen or tamoxifen (6). Over the years, his research expanded to other steroid receptors, including the androgen receptor, which is a critical target in prostate cancer.

Discovering Coregulators

Brown stayed in Sharp’s laboratory until 1990. By 1991, he was back at Dana–Farber Cancer Institute, forming his own laboratory and teaching at Harvard Medical School. Through the 1990s, he began chipping away at understanding the workings of the ER. Its main function, explains Brown, is to turn on genes in response to estrogen. “What happens in breast cancer,” he says, “is a change in the function of the ER that not only leads to appropriate growth and differentiation but leads to inappropriate growth.”

Brown and his colleagues discovered that there are other proteins that affect whether and when genes are turned on or off, along with estrogen and the ER. These proteins are called coregulators, both coactivators and corepressors, and Brown and coworkers (7) discovered that certain coactivators bind to the ER only when it is activated by estrogen. Subsequent work has shown that once these coregulators bind to the ER, they bring the transcriptional machinery to the genes that the ER binds to.

“Without the coactivators, you wouldn’t get as robust transcription,” explains Brown. “They are there to amplify and integrate signals from other pathways in quite a complicated way that we still can’t completely describe.”

After Brown’s discovery, other researchers started finding different coregulatory proteins that interact with the ER and other steroid receptors. Many assumed that these coregulators bind in no particular order. However, in a seminal paper published in Cell in 2000, Brown and coworkers (8) found that the coregulators assembled and disassembled in a systematic way. Understanding the sequence can help researchers target therapeutic pathways in cancer, says Brown.

Along with understanding the steps by which the ER and its coregulators assemble, Brown and his colleagues also needed to determine where in the genome they bind to trigger transcription. To help with the computational complexity of this work, Brown teamed up with Dana–Farber Cancer Institute computational biologist Shirley Liu, who developed powerful tools. In an influential paper (9), Brown, Liu, and their coworkers showed that ER does not always bind to the proximal promoter of the genes it regulates. Instead, it binds more frequently at sites distant from the gene, even hundreds of kilobases away, where it acts together with the transcription factor FOXA1 to form estrogen-regulated enhancers. Later, Brown and Liu would coin the term “cistrome” to describe all of a transcription factor’s binding sites in the genome (10). So, the ER cistrome would be all of the places on a genome the ER binds. Understanding cistromes, Brown says, provides insight into which genes ER regulates in a given cell type, such as breast cells.

“You can imagine that if you knew that the ER, together with some other factor, bind[s] to a distinct set of regulatory regions that were only active in breast cells, then you might be able to come up with a way to target those specifically,” says Brown.

Mapping Genetic Feedback Loops

Brown and Liu continue to piece together the complex interactions between hormones such as estrogen and receptors, coregulators, and genes. Through it all, Brown has continued to ponder the question central to his work: Why do drugs like tamoxifen sometimes stop working? Brown is a step closer to the answer with his Inaugural Article (1), which uses CRISPR to conduct a comprehensive genetic screen to identify the genes that affect the survival of ER-positive breast cancer cells.

“What was gratifying was that the top genes that we found to be essential for the estrogen-stimulated growth of ER-positive breast cancers were ones we had been identifying over the last couple of decades—transcription factors like FOXA1 and GATA3 and SPDEF, which are part of the breast-cell identity program,” says Brown.

Even more importantly, the team found a gene, called c-terminal SRC kinase (CSK), that, when knocked out, boosted the growth of cancer cells in the absence of estrogen. Further analysis showed that CSK is a critical player in a previously unknown negative feedback loop where estrogen stimulates the ER to turn on growth genes but also turns on CSK, which inhibits SRC family members and limits how strongly estrogen stimulates cell growth. So, while estrogen deprivation or ER antagonists like tamoxifen dampen estrogen-fueled cell growth, they turn on SRC signaling normally inhibited by CSK. In turn, SRC signaling can reactivate the ER and turn on other growth-promoting signals, says Brown.

“That’s an unfortunate consequence of inhibiting the ER,” says Brown. “By interrupting this feedback loop, you activate this other signaling pathway.”

Fortunately, additional screens allowed Brown’s laboratory to identify another gene, PAK2, downstream from SRC family signaling that becomes essential in the absence of CSK. If the team inhibited that gene as well as ER, they could more fully inhibit the growth of ER-positive tumor cells, essentially blocking the ER and downstream reactivation.

Brown hopes that his work will lead pharmaceutical companies to develop more effective therapies for breast cancer. His team is already testing whether they can develop PAK2 inhibitors that may be paired with ER-targeting therapies in breast cancer. “I believe it will be these kinds of multipronged approaches that will eventually lead to better treatments,” he says.

Overall, Brown believes his research program demonstrates the importance of advances such as CRISPR screening to disease treatment. He adds that “it takes the combined efforts of curiosity-driven basic researchers and disease-oriented researchers” to provide the foundation for advances in medicine.