Maintenance of Constitutive and Inactive X Heterochromatin in Cancer and a Link to BRCA1: A Dissertation

Publication Date

June 2007

Document Type

Doctoral Dissertation


Graduate School of Biomedical Sciences, PhD/MD Program


BRCA1 Protein; Chromosomes, Human, X; Heterochromatin; RNA, Untranslated; Centromere; DNA Replication; Protein-Serine-Threonine Kinases; Academic Dissertations; Dissertations, UMMS


The development of cancer is a multi-step process which involves a series of events, including activation of oncogenes and loss of tumor suppressor function, leading to cell immortalization and misregulated proliferation. In the last few years, the importance of epigenetic defects in cancer development has become increasingly recognized. While most epigenetic studies focus on silencing of tumor suppressors, this thesis addresses defects in the maintenance of silenced heterochromatin in cancer, particularly breast cancer. Breast cancer is a leading cause of cancer in women and many familial cases have been linked to mutations in the breast cancer susceptibility genes, BRCA1 and BRCA2. BRCA1 has been linked to DNA repair as well as multiple other cellular processes, including cell cycle checkpoints, ubiquitination, centrosome function, and meiotic silencing of the XY body. This work began with a particular interest in the report that BRCA1 was linked to the failed maintenance of random X-inactivation in female somatic cells, via a role in supporting XIST RNA localization to the inactive X chromosome (Xi). XIST RNA is a non-coding RNA that fully coats or “paints” the Xi and induces its silencing. Work presented in Chapter II substantially clarifies the relationship of BRCA1 to XIST RNA, based on several lines of experimentation. Loss of BRCA1 does not lead to loss of XIST RNA in these studies, nor did reconstitution of HCC1937 BRCA1-/- tumor cells with BRCA1 lead to XIST RNA localization on Xi, although an effect on XIST RNA transcription is possible. Studies of BRCA1 localization with Xi showed that BRCA1 has a limited association with the Xi in ~3-10% of cells, it rarely colocalizes with XIST RNA to a significant extent, but rather is in close apposition to a small part of the XIST RNA/Xi territory. Additionally, analysis of several breast cancer cell lines revealed mislocalization of XIST RNA in some breast cancer cell lines.

Many studies have examined BRCA1 foci that form following DNA damage and demonstrated that these are sites of repair. However, whether the numerous large foci consistently present in normal S-phase nuclei were storage sites or had any function was unknown. In Chapter III, I demonstrate that the BRCA1 foci in normal S-phase nuclei associate overwhelmingly with specific heterochromatic regions of the genome. More specifically, BRCA1 foci often associate with centromeric or pericentromeric regions in both human and mouse cells. In human cells BRCA1 foci often appear juxtaposed to centromeric signal, whereas in mouse, BRCA1 often rings or paints the large chromocenters, clusters of DAPI-dense pericentric and centric heterochromatin. Using PCNA and BrdU as markers of replication, I demonstrate that BRCA1 preferentially associates with the chromocenters during their replication, although high-resolution analysis indicates that BRCA1 and PCNA foci rarely directly overlap. Interestingly, cells with defects in BRCA1 were found to have lagging chromosomes and DNA bridges which nearly always contained satellite DNA, which is consistent with the possibility that BRCA1 deficit contributes to failed separation of sister chromatids at the centromere. This is consistent with other recent reports that BRCA1 is necessary for DNA decatenation by topoisomerase II during routine replication and with my demonstration that topoisomerase II also accumulates on pericentric heterochromatin (PCH) during replication.

Chapter IV presents recent work which reveals that RNA is commonly expressed from the centric/pericentric heterochromatin and appears to be linked to its replication. In mouse cells RNA from heterochromatic sequences is readily detected using a broad molecular cytological assay for repeat transcription (the COT-1 RNA assay). In addition to a more dispersed nucleoplasmic signal from euchromatic nuclear regions, distinct localized foci of repeat RNA are detected with COT1 probe or pancentromeric probe. Further analysis with the minor satellite (centromere proper) and the major satellite (comprising the larger pericentric heterochromatin) reveals that the large RNA foci often contain these satellite sequences, long thought to be essentially silent. These foci generally associate with the PCH of chromocenters, and produce various patterns similar to BRCA1- including a larger signal partially painting or ringing the chromocenter in a fraction of cells. In conjunction again with PCNA staining, it was possible to determine that the major satellite RNAs associate with the chromocenters during replication. While the satellite RNA co-localizes precisely with PCNA, neither of these co-localizes at high resolution with BRCA1, although they all are present on replicating chromocenters contemporaneously. These findings show that satellite RNAs are more widely expressed in normal cells than previously thought and link their expression to replication of centromere-linked heterochromatin.

Finally, Chapter V presents three lines of recent results to support a major concept forwarded in this manuscript: that loss of Xi heterochromatin may reflect defects in the broader heterochromatic compartment, which may be manifest at multiple levels. I provide evidence using two new assays that both the peripheral heterochromatic compartment and the expression and silencing of satellite repeats is commonly compromised in cancer, although this appears to vary among cancer lines or types. The final results connect back to the question with which I began: what maintains XIST RNA localization to the chromosome in normal cells. These results demonstrate for the first time that Aurora B Kinase activity, mediated by Protein Phosphatase 1 (PP1) during interphase, controls the interphase retention and mitotic release of XIST RNA from the chromosome, likely linked to chromatin modifications such as H3Ser10 phosphorylation. As Aurora B Kinase is commonly over-expressed in cancer and is linked to chromatin changes, this exemplifies one type of mechanism whereby broad epigenetic changes in cancer may impact XIST RNA localization and the maintenance of heterochromatin more generally. This thesis represents a melding of cancer biology with the study of X inactivation and heterochromatin, with findings of fundamental interest to both of these fields.


In the process of seeking author's permission to provide full text.

Rights and Permissions

Copyright is held by the author, with all rights reserved.

This document is currently not available here.