A cluster of breast cancer cells undergoing programmed cell death WELLCOME IMAGES, ANNIE CAVANAGH

The discovery of the blockbuster tumor suppressor gene breast cancer 1, early onset (BRCA1) in the early 1990s was a major breakthrough in unraveling the genetics of common hereditary cancers such as breast and ovarian. Yet, after nearly two decades of research, mystery still surrounds the mechanism BRCA1’s protein product employs to suppress tumor growth.

In a new study published today (October 27) in Science, researchers pegged this crucial function to a specific region of the protein known as the BRCT domains: two nearly identical stretches, around 90 amino acids long, that lie toward the carboxyl-end of the protein and are responsible for binding phosphorylated proteins.

“It’s a very important study,” said Steve Smerdon, a structural biologist at UK’s National Institute for Medical Research, who was not involved in the research. “It’s...

Since its discovery, BRCA1 has been associated with many cellular processes, including transcriptional regulation, chromatin remodeling, cell cycle control, and repair of double strand DNA breaks.

Researchers have linked more than 1,500 mutations in BRCA1 to an increased risk of developing early onset cancer in humans. While the majority of these are frameshift and nonsense mutations that result in a severely truncated and non-functional protein, a minority are point mutations that substitute a single amino acid in the 1,863-amino-acid-long BRCA1 protein.

Researchers have noticed that many of these point mutations cluster in two regions of BRCA1: the BRCT domains in the carboxyl end of the protein, and a region about 110 amino acids long on the opposite end of the protein known as the RING domain. The latter is known to bind to the protein BARD1 (BRCA1-associated RING domain protein 1) and form a complex that helps attach ubiquitin to proteins.

Structure of the BRCA1 protein
Structure of the BRCA1 protein
PROTEIN DATA BANK

While both the BRCT and RING domains have been implicated in tumor suppression, Columbia University cancer researchers Thomas Ludwig and Richard Baer, who led the current study, wanted to definitively find out if it was the binding of phosphorylated proteins by the BRCT domains, the ubiquitinating capability of the RING domain, or both, which were helping to keep cancer at bay.

To investigate this, the researchers engineered mice in which one of the two copies of the BRCA1 gene harbored a mutation that replaced an isoleucine in the RING domain with an alanine. While this mutation still allowed BRCA1 to bind to BARD1 (which is necessary for the stability of both proteins), it impaired the complex’s enzymatic ability to ubiquitinate.

The researchers then deleted the normal allele in the epithelial mammary glands of the mice—thus the cells in this tissue carried only the mutated BRCA1. This procedure modeled what happens in humans with BRCA1 mutations, who always carry one good copy of the gene, but who presumably develop tumors from cells that lose the good copy.

To their surprise, they found that these mice were no different than wildtype mice. They did not develop tumors or display any aberrant growths. In contrast, mice with a mutation that grossly deforms BRCA1 develop tumors 15 months after their good BRCA1 copy is deleted.

Moreover, when the good BRCA1 copy was knocked out in the pancreas of mice carrying the BRCA1 mutation in the RING domain in addition to cancer-causing mutations in the genes p53 and Kras, the mice developed tumors with the same speed and pathology as mice with mutations in p53 and Kras only—suggesting that lack of the ubiquitin ligase activity of the RING domain played no part in tumorigenesis.

“We found we do not need the enzymatic activity [of the RING domain] in cells for any of the functions attributed to BRCA1 which was very surprising and totally unexpected,” Ludwig said.

Curiously, male mice homozygous for the BRCA1 RING domain mutation were sterile. “So there clearly is an important function for the ubiquitin ligase, at least during spermatogenesis,” Ludwig added.

But mice with a point mutation in the BRCT domain, which impaired BRCA1’s ability to bind the phosphate group of phosphorylated proteins, did develop tumors in the mammary glands and the pancreas after the wildtype BRCA1 allele was knocked down.

“These mice develop tumors almost as rapidly as mice with null mutations in BRCA1,” Baer said, suggesting that the binding of BRCA1 to other phosphorylated proteins is critical for tumor suppression.

Indeed, proteins known to bind to BRCA1’s BRCT domains—such as Abraxas, BACH1, and CtIPare all involved in the DNA repair process. This meshes with the prevailing view that faulty DNA repair by a mutated BRCA1 results in genomic instability, which ultimately leads to tumorigenesis.

But that is not to say that the RING domain of BRCA1 is not important, said Mark Glover, a structural biologist at the University of Alberta who was not involved in the study. Although the domain’s ubiquitin ligase activity is still shrouded in mystery and does not seem to be involved in tumor suppression, the RING domain plays a crucial role in binding BRCA1 to BARD1, an interaction that is known to function in tumor suppression as well.

“Given the number of people that are interested in BRCA1 ranging from clinicians all the way to structural biologists, this study is going to have a huge impact in understanding not only BRCA1 tumor suppression functions,” Smerdon said, “but also it will probably will give us some insights into how other tumor suppressors known to act alongside BRCA1 are functioning at the molecular level.”

R. Shakya, et. al., “BRCA1 tumor suppression depends on BRCT phosphoprotein binding, but not its E3 ligase activity,” Science, 525-8, 2011.

Interested in reading more?

Become a Member of

Receive full access to more than 35 years of archives, as well as TS Digest, digital editions of The Scientist, feature stories, and much more!