Cellular, epigenitic, genetic and signalling alterations associated with RANK expression in bone-tropic breast cancer cells

Abstract

Bone metastases are a major cause of morbidity in patients of advanced breast cancer. Development of osteolytic bone metastasis depends on the interaction between malignant tumour cells and bone microenvironment. Receptor Activator of Nuclear Factor Kappa B (RANK) is a member of tumour necrosis factor (TNF) superfamily that is expressed by osteoclasts (the bone resorbing cells) and primary breast tumour cells. Previous studies demonstrated that RANK receptor and its ligand (RANKL) play an important role in bone remodelling, mammary gland development and immune system. RANKL was also found to serve as a chemotactic factor that facilitates breast tumour metastasis to bone. However, the role of the RANK receptor in breast cancer cell metastatic behaviour in bone is not fully understood. Therefore, the aim of this thesis was to explore the role of the RANK receptor in parental and bone-tropic breast cancer cell growth, motility and invasion, and assess these cells influence on breast cancer cell induced osteoclastogenesis. Functional studies in breast cancer cells showed that RANKL (100 - 300 ng/ ml) significantly enhanced parental human MDA-231 (MDA-231P) and mouse 4T1 breast cancer cell spreading within minutes. RANKL induced chemotactic cell migration of MDA-231P cells in vitro. I also found that RANKL significantly stimulated random and directional 2D and 3D cell migration of parental MDA-231P and bone-tropic (MDA-231BT) breast cancer cells in vitro. These effects were observed at concentrations (100 – 300 ng/ml) that were sufficient to induce osteoclast formation in the presence and absence of breast cancer cells in vitro. In contrast, high concentrations of RANKL (1000 ng/ ml) dramatically suppressed human MDA-231P breast cancer cell invasion in vitro. These data indicate that the RANK receptor in the breast cancer cell lines tested influences cancer cell spreading, migration and invasion in vitro. Thus, targeting RANK in tumour cells may be of value in the prevention of tumour burden associated with breast cancer bone metastasis. Mechanistic studies revealed that RANKL stimulated the phosphorylation of p38 kinase in human and mouse breast cancer cells. Interestingly, RANKL had no effect on NFᴋB, JNK and AKT pathways in parental human MDA-231 and mouse 4T1 breast cancer cells at concentrations up to 300 ng/ ml. These data implies that the RANK receptor modulates human and mouse breast cancer cell metastatic behaviour via p38 activation and independently of the NFᴋB and PI3K/AKT pathways. Silencing of the RANK receptor in the bone-tropic human breast cancer cells MDA- 231BT2 reduced directional cell migration without affecting cell viability and growth. Functional studies in osteoclast and breast cancer cell revealed that knockdown of RANK expression in both parental and bone-tropic human breast cancer cells significantly inhibited the ability of these cells to stimulate osteoclast formation. Although, I cannot exclude the possibility of the involvement of other signalling pathways downstream of the RANK receptor, these studies suggest that the RANK/P38 signalling in osteoclast and breast cancer cells contributes significantly to breast cancer cell behaviour in bone. Genetic analysis of the RANK gene in human parental and bone-tropic MDA-231 breast cancer cells showed a number of polymorphisms. One variant detected was found to be deleterious for the RANK protein. This variant changes the amino acid sequence from alanine to threonine (Ala ˃ Thr) and only appeared in the RANK gene in the parental human MDA-231P breast cancer cells. Moreover, of the four known RANK isoforms that were detected in the parental and bone-tropic breast cancer cells tested, two lacked the TRAF6 binding motifs associated with NFκB activation. All RANK isoforms detected on the bone-tropic MDA-231BT breast cancer cells expressed the P38 binding motifs. Altogether, these findings support the role of the RANK/P38 signalling pathway in breast cancer cell behaviour in bone. Epigenetic analysis in parental human MDA-231P breast cancer cells showed that continuous and long-term exposure to RANKL (10 and 100 ng/ ml) for up to 50 passages (approximately 120 days) did not induce epigenetic changes, particularly DNA methylation, in the RANK gene. However, I found DNA methylation changes in a set of genes that are known to be involved in cell development and regulation. The methylation status of the altered CpG loci either hypermethylated or hypomethylated are located at different parts in the CpG islands. Whole genome DNA methylation pattern of the bone-tropic breast cancer cells showed a number of genes that appeared in both bone-tropic variants are correlated with different biological function of the cells. I also found that long-term exposure of human MDA-231P to RANKL (100 ng/ ml) enhanced the ability of these cells to stimulate osteoclastogenesis in vitro. These data together indicate that long-term exposure to RANKL induces “boney” epigenetic changes in a set of genes that enhances breast cancer cell behaviour in bone. Overall, this thesis illustrated that the RANK receptor on human parental and bone-tropic breast cancer cells plays an important role in cell motility and ability of these cells to influence osteoclastogenesis and ultimately osteolysis. Therefore, agents that selectively target the RANK receptor may be of value in the treatment of both tumour burden and osteolytic bone disease associated with breast cancer. However, the role of the RANK receptor in bone metastasis will require further in vivo investigation.