Using DNA methylation to study benign breast and breast tumours

© 2011 Dr. Katie Tzu-Ai HuangBreast cancer is a heterogeneous disease comprising of numerous distinct entities that have different biological and clinical behaviours. Clinical diagnosis and accurate prognosis are not infrequently difficult to make, leading to over or under treatment. Although improvements in survival have been observed over the last 10 years, a substantial number of women die from the disease. A better understanding of the molecular mechanisms and signalling pathways that are involved in breast cancer development (e.g. the hypoxia pathway) as well as a deeper appreciation of the role of epigenetic alterations through DNA methylation, will greatly improve our combative efforts against breast cancer. The aim of this thesis is to assess DNA methylation of candidate genes in breast cancer with the goal of identifying biologically important pathways of hypoxia-inducible factor 1 (HIF-1) and to find novel diagnostic and prognostic markers. HIF-1 is the master regulator of cellular oxygen homeostasis that is induced by hypoxia. DNA methylation assessment was performed on the genes that regulate HIF-1 to understand the link between DNA methylation changes in cancer and the altered response to hypoxia. However, promoter methylation of a panel of HIF-1α regulating genes (VHL, CUL2, TCEB1, TCEB2, RBX1, PHD1, PHD2, PHD3, FIH, SDHA, SDHB, SDHC, SDHD, FH and SIAH2) was not present in breast cancers. This showed that promoter methylation is unlikely to be an important mechanism in stabilising HIF-1 in breast cancers through the down-regulation of the expression of these genes. Nevertheless, the study has demonstrated that the loss of expression of CITED4, a negative regulator of HIF-1, in breast cancer is in some cases due to DNA methylation that leads to increased HIF-1α activity and its downstream genes. Demethylation and histone modification was able to reactivate CITED4 gene expression in the methylated cell lines and lead to changes in tumour behaviour. The data of the study also suggests that histone modification may be the alternative epigenetic mechanism that regulates CITED4 expression. The distinction between fibroadenomas and benign phyllodes, which are both fibroepithelial tumours of the breast, can be difficult with no specific features that can reliably distinguish these neoplasms. I have demonstrated that the strong methylation of RASSF1A and/or TWIST1 was associated with phyllodes tumours. TWIST1 methylation is also correlated with increasing malignancy in phyllodes tumours. Hence, RASSF1A and TWIST1 methylation, and potentially other markers such as APC and WIF1, may be useful clinically for distinguishing phyllodes tumours from fibroadenomas. Additionally, the absence of frequent methylation in fibroadenomas supports a non-neoplastic origin. DNA methylation profiling and array comparative genomic hybridisation have been demonstrated as a valuable diagnostic tool in differentiating newly arisen tumours for ‘true’ relapses from new independent tumours. DNA methylation profiling results showed that ipsilateral tumours have a higher chance to be a recurrent tumour, whereas contralateral tumours were mostly a de novo tumour. Overall, this thesis have demonstrated that DNA methylation is an important mechanism in understanding the depth of hypoxia in breast cancer and has the potential to be a useful diagnostic tool in helping to answer some clinical questions in breast cancer