Research Grants Awarded
X chromosome contribution to basal-like breast cancers
Breast cancers arising in women who carry a BRCA1 germline mutation cluster with the sporadic basal-like subgroup of breast cancers (BLCs) by both gene expression analysis and immunocytochemical phenotyping. Hence, there is the possibility that BLCs may phenocopy BRCA1-deficient tumors by sharing defects in common molecular pathways. We have previously shown BRCA1-deficient tumor cells to have defects in the heterochromatin superstructure that is associated with the inactive X chromosome (Xi). Specifically, both BRCA1-deficient human cell lines and human and murine BRCA1-mutated tumors were shown to lack an X chromosome decorated with XIST RNA, macrohistone H2A 1.2 (MH2A), and histone H3 methylated on Lysine 27 (H3mK27), all established features of the facultative heterochromatin of the Xi. Given the phenotypic similarities between BRCA1-deficient and sporadic BLCs, we have asked whether sporadic BLCs also display X chromosomal abnormalities. We performed immunofluorescence staining and detected proper expression and multifocal subnuclear localization of BRCA1 in sporadic BLC (and sporadic non-BLC, used as controls) tumor samples. Mutational analysis confirmed both groups of tumors to be genetically wild type for BRCA1. However, analysis of frozen sections revealed that in tumor cells of sporadic BLCs there is almost always a failure of localization of XIST RNA, MH2A and mK27H3 on either X chromosome (Livingston Lab, unpublished). Importantly, DNA FISH and SNP array analysis confirmed that these XIST-negative, basal-like tumors contain at least two X chromosomes, suggesting that there is a specific defect in the maintenance of Xi. Further investigation using high-density SNP arrays has demonstrated that approximately half the basal-like specimens display clear evidence of X chromosome isodisomy (likely due to loss of Xi and duplication of the active X chromosome). ~40% of sporadic BLCs analyzed contained either bi-parental but non-heterochromatinized X chromosomes or gains of additional X chromosomal DNA. In contrast, only 10% of non-BLCs displayed these particular X chromosome changes. The nearly uniform presence of an Xi defect and the known role of BRCA1 in promoting proper Xist/Xi localization strongly suggest that misbehavior at the inactive X chromosome plays a role in the pathogenesis of the BLC subset of breast cancers, both BRCA1 -/- and sporadic. However, how such a chromosomal defect contributes to the tumor development process is unclear. Whether, for instance, the absence of heterochromatic features translates into reawakening of previously silenced genes in the subgroup with retention of biparental X chromosomes or corresponds to the presence of two transcriptionally active identical alleles in the subgroup with isodisomy, is still unknown. The detected overexpression of a small number of genes clustered in distinctive X chromosomal regions (Livingston lab, unpublished) is consistent with such speculation. The primary goal of my work will be to learn how the observed X chromosome heterochromatic defects contribute to the pathogenesis of BLCs. It can be hypothesized, for example, that the above-noted X chromosome-associated abnormalities lead to biallelic expression of certain, normally monoallelically expressed, X chromosomal genes. My analysis will be, initially, focused on the small subset of X chromosome genes found to be overexpressed in sporadic and BRCA1 -/- BLCs. Multiple approaches, including nascent RNA FISH, will be used to discern whether these genes are actively expressed from one or both X chromosomes. If any of the genes is consistently characterized by biallelic expression, then it will be important to try to discern whether it contributes to the tumor development process. The identification of a gene that contributes to the viability of BLC tumor cells would elicit immediate interest in whether it could serve as a viable BLC drug target.
Breast cancer accounts for ~ 32% of the total number of new cancer cases diagnosed in American women each year. Up to 10% of breast cancer cases are due to underlying familial predisposition. BRCA1 and BRCA2 are the two major predisposing genes identified so far. Women who inherited from either parent a mutation in the BRCA1 gene have a 50-80% lifetime risk to develop a specific (basaloid) subtype of breast cancer. BRCA1-linked tumors share genetic and clinical features with an aggressive subgroup of sporadic breast cancers, termed basal-like cancers (BLC). BLC currently lack effective or life prolonging therapy beyond surgery and local radiation therapy. An essential function of the BRCA1 protein is to defend cells from the deleterious effects of DNA damage. We have previously shown that BRCA1 is also required in the process that normally leads to the inactivation of the X chromosome in the cells of women. Such a female-specific process renders all cells of a female (XX) organism equal to the cells of a male (XY) individual, in terms of number of active X chromosomes, by ‘shutting down’ the functions of one of the two X chromosomes. The ‘silenced’ X chromosome displays unique marks of its inactive status. The Livingston laboratory has shown that typical features of the inactive X are absent in hereditary BRCA1-mutated tumors. More recently, we found also sporadic basaloid breast cancers to be characterized by the same X chromosomal defects. X chromosomal abnormalities seem to be a specific feature of BLC, both inherited (i.e. BRCA1 -/-) and sporadic, suggesting that such defects contribute to the emergence of this malignancy. The aim of my study will be to investigate if and how the ‘misbehavior’ of the X chromosome, as detected in the tumors analyzed, plays a role in the development of the BLC subtype of breast cancers. The following questions will be addressed in my research plan: 1) does the loss of typical features of the inactive X result in the presence of two, transcriptionally active copies of X chromosome? 2) if so, are all silenced X genes reawakened and, if not, which are? 3) are the same X genes reawakened in a sizeable number of different sporadic (BRCA1 wt) and BRCA1 -/- BLC samples (suggesting that some of them actually contribute to BLC tumorigenesis)? If contribution of an identified gene to the emergence of basaloid tumors can be established, it would become a provisional target for directed experimental therapy.