Mutations in the tumour suppressor gene BRCA1 are associated with a greatly elevated risk of developing breast and ovarian cancer, thought to be as high as 95% by the age of 70. Although much is now known about the expression and functions of this gene – it has been implicated in a wide range of processes including cell-cycle checkpoint activation, transcriptional regulation and DNA damage repair – no mechanism of action is yet known that can fully explain its role in these disparate processes. However, BRCA1 is known to be expressed at high levels in sites in the brain that are associated with neurogenesis.

A group of researchers led by Gerald M. Pao and Inder M. Verma at the Salk Institute for Biological Studies, La Jolla, California, USA, have investigated the molecular mechanisms of the BRCA1 protein, initially through studying the brains of transgenic mice lacking the mouse ortholog of this gene, Brca1. The researchers found using microarray analysis that epigenetic regulation was disrupted in the brains of Brca1 knockout mice, with de-regulation of some imprinted genes. Condensed regions of heterochromatin containing repetitive tandem DNA, which are known as heterochromatic centres and are particularly visible in neurons, were significantly reduced in number but larger and more diffuse in Brca1-/- mouse brains compared to controls. BRCA1 has previously been shown to ubiquitinylate a histone known as H2A, which is one of five main histones in the structure of chromatin in eukaryotic cells, and the researchers also observed that the heterochromatic centres in knockout mice were deficient in ubiquitinylated H2A. These observations, taken together, provided Pao and Verma with the first suggestion of a link between BRCA1 and the organisation of heterochromatin structure.

Condensed regions of DNA are associated with transcriptional silencing. The researchers next investigated the transcriptional status of the DNA in the defective heterochromatic centres, and found that mRNA transcripts of these regions of satellite DNA were elevated in the mutant brains compared to controls. Similar disruption to heterochromatin structure and function was observed in the mammary glands of young female Brca1-/- mice. They then investigated the relationship between BRCA1, histone H2A and heterochromatin in vivo using chromatin immunoprecipitation (ChIP), and showed that ubiquitinylated H2A is enriched at satellite DNA regions compared to other regions of repetitive DNA. This indicates that BRCA1 acts as a ubiquitin ligase for histone H2A at these regions. These results and other investigations with distinct BRCA1 mutants confirmed the association of ubiquitin ligase activity with a specific domain of the BRCA1 protein, the RING finger.

Next, Pao, Verma and their co-workers generated a construct of ubiquitin fused to histone H2A that can mimic the natural ubiquitinylated histone. When BRCA1-deficient human breast cancer cells were transduced with a lentivirus expressing this construct, expression of DNA in satellite repeat regions was reduced to the levels seen in wild type cells. In Brca1-/- mouse cells, too, expression of a H2A-ubiquitin construct was shown to restore transcriptional silencing of satellite DNA and reduce apoptosis. Furthermore, inducing the over-expression of satellite RNA in human mammary epithelial cells generated a pattern of DNA defects, including double-strand breaks, that mirrors those seen in BRCA1-deficient cancer cells. This suggests that the satellite DNA over-expression that results from a lack of histone H2A ubiquitinylation is a cause of the genomic instability that leads to cancer rather than simply a peripheral effect.

Taken together, these results indicate that the mechanism through which BRCA1 maintains the stability of genomic DNA is through ubiquitinylation of histone H2. Ubiquitinylated H2 preferentially binds to heterochromatic satellite DNA, and this prevents its over-expression. This elucidation of the mechanism through which this protein carries out many of its tumour suppressor activities may lead to the development of novel therapies for BRCA1-deficient tumours.

Reference

Zhu, Q., Pao, G.M., Huynh, A.M., Suh, H., Tonnu, N., Nederlof, P.M., Gage, F.H. and Verma, I.M. (2011). BRCA1 tumour suppression occurs via heterochromatin-mediated silencing. Nature 477, 177-184. doi:10.1038/nature10371