Studies Chip Away at Sex Hormone Roles in Prostate and Breast Cancers

If you ask parents for advice on keeping an adolescent out of trouble, they might report the benefits of repeating, as often as possible, just three questions: Where are you? Who are you with? What are you doing?

For some cancer-related proteins, these same questions are highly relevant, especially when (just as for teens) there are hormones involved. In the case of breast cancer and of prostate cancer, the actions of steroid sex hormones and their receptor proteins are inseparable from the initiation, growth, progression, and prognosis of disease.

Hormone-induced chromatin looping. Androgen receptor (AR) binding to enhancer and promoter domains of the gene for prostate specific antigen triggers the bending of DNA, assembly of cofactors (unmarked ovals), and tracking of RNA polymerase (Pol II) to the gene transcription start site (TATA).

Checking Up
In prostate cancer, androgens and their receptor run the show, while estrogens and the estrogen receptor are the culprits in breast cancer. Both kinds of receptor share the common ability to bind to DNA and regulate gene activity, and both are important targets for anticancer treatments.

But these same hormones that boost the expression of cancer genes also control genes that maintain healthy tissues like bone, muscle, and sex organs. Improving antihormone cancer therapies, then, means finding ways to steer steroid receptors away from the cancer-causing genes and toward the genes that promote health. To reach this goal, HMS associate professor of medicine Myles Brown has dedicated his lab at the Dana–Farber Cancer Institute to an intensive experimental interrogation of both the estrogen and androgen receptors, with the intention of finding out as much as possible about where they are and with whom, and how they are behaving.

The Brown lab’s recent findings give the first detailed account of the androgen receptor’s actions as it regulates the gene for the important cancer marker prostate specific antigen (PSA). By showing the androgen receptor at work, binding and bending DNA and cooperating with accessory factors, the research uncovers some general aspects of gene regulation by steroid hormones and also some specific features of how the androgen receptor interacts with the PSA gene. “This work advances our basic understanding of one example of how the androgen receptor regulates a recognized target gene and forms a basis for us to go further to understand how it regulates all of its target genes,” said Brown.

Photo by Steve Gilbert

Myles Brown (right) and research fellows Jason Carroll (left) and Qianben Wang pursue better treatments for prostate and breast cancer by probing androgen and estrogen receptor dynamics.

In the work, published in the Sept. 2 Molecular Cell, first author Qianben Wang used a method called chromatin immunoprecipitation (ChIP) to isolate androgen receptor along with associated gene fragments from hormone-treated cells. By checking the tag-along DNA for sequences that came from near the PSA gene, Wang showed that over the course of 16 to 24 hours after hormone treatment, the androgen receptor gradually latched onto two spots on DNA upstream of the PSA gene, one close to the gene and one about 4,000 base pairs away. These two regions, the promoter and the enhancer, were known to cooperate in recruiting the RNA polymerase complex to the gene and start transcription, but how they acted together while being physically distant on the linear chromosome was not understood.

A modification of ChIP that allows analysis of the three-dimensional structure of the gene bound to the androgen receptor gave the answer: the androgen receptor proteins on the enhancer and promoter were, in fact, drawn together, causing the intervening DNA to create a loop structure. The researchers found RNA polymerase all along the loop, suggesting that the enzyme got onto the DNA at the enhancer and traveled down the loop like a train on a track, passing the promoter sequence to finally reach the PSA gene itself and begin transcription. In addition to revealing both the kinetics and the mechanism of how the androgen receptor promoter–enhancer pair jump-starts PSA transcription, Wang also identified the important accessory proteins for this process.

“This work advances our basic understanding of one example of how the androgen receptor regulates a recognized target gene and forms a basis for us to go further to understand how it regulates all of its target genes.”

Knowing just how the androgen receptor turns on PSA is only the first step to better treatments for prostate cancer, according to Brown. “The PSA gene is a good indicator of the presence of prostate cancer cells, but what we really need to know is what are the important genes that are responsible for cell proliferation, and what are the genes that are responsible for the more physiological effects of androgens,” he explained. “Surprisingly, few of the genes that are direct targets of the androgen receptor are known.”

Chips Ahoy
To move from a single gene to a somewhat bigger picture, Brown is collaborating with DFCI colleagues Pamela Silver and Shirley Liu to generate a genomewide map of all the androgen-binding sites in human cancer cells treated with hormones. By combining the ChIP method of isolating the androgen or estrogen receptor with a DNA microarray chip to analyze the gene sequences that the receptors carry along with them, the researchers can assemble a complete picture of where the steroid hormone receptors are and what genes they are regulating. In a paper published in Cell on July 15, research associate Jason Carroll and collaborators from the Silver and Liu groups reported their first analysis, this one of estrogen receptor binding sites on human chromosomes 21 and 22. Brown reports the group is now putting the final touches on their map of androgen receptor binding sites on chromosomes 21 and 22 and is extending the analysis to cover the entire human genome for both receptors.

Among the surprises the researchers got from their estrogen receptor experiments, they found relatively few estrogen-responsive enhancer sites on the two chromosomes and discovered some that were much farther from genes than expected. In many cases, the enhancers would never have been identified by traditional methods of promoter analysis, which most often rely on cloning and testing of gene fragments under artificial conditions. “The ChIP on chip method is an extremely useful and powerful tool because it lets us look at steroid–receptor function on endogenous genes in real cells,” Brown said. With that information in hand, the researchers are confident they can do more than just chip away at finding better inhibitors of steroid receptors to treat breast and prostate cancers.

—Pat McCaffrey

top