Metastatic dissemination is the major cause of cancer associated death. Dissemination of tumor cells is a very early event in cancer progression and metastatic cells relapse even decades after the first diagnose and can show different clinical features compared to the primary disease. In breast cancer, bone, lung and brain are the major sites of tumor metastases. The “seed and the soil” theory propose that selected population of cells can spread from the primary tumor in different organs based on different requirements. Moreover, specific environments can differently influence cancer cells and affect response to therapy. For these reason it is important to investigate the role of the tumor microenvironment and to dissect the molecular cross-talk between cancer cells and the surrounding cells in the stroma.
Bone marrow (BM) is particularly permissive for cancer dissemination thanks to environmental and structural features like the fenestration of bone sinusoidal capillaries that facilitates the entrance of circulating cells. Moreover, BM is a key reservoir for metastatic breast cancer cells. By still unknown mechanisms, tumor cells can enter into a quiescent state in the bone and remain dormant for very long periods. Regarding this, an important issue is that dormant tumors show resistance to conventional therapy that cannot target non-proliferating cells.
When tumor relapse occurs cancer cells reacquire the proliferative ability, thereby causing osteolytic lesions. During this process cancer cells secrete factors that increase recruitment and activation of osteoclasts, driving osteolysis. In turn, matrix resorption causes the release of soluble factors that sustain tumor cell proliferation and promote the progression of bone metastases, establishing the so called “vicious cycle”. Pre-clinical and clinical evidences suggest that to hit the vicious cycle is a therapeutic strategy, since agents that inhibit bone resorption, like bisphosphonates, inhibit tumor cells expansion.
We aimed to analyze molecular mechanisms involved in the interaction between tumor cells and the BM environment to identify suitable targets in the tumor cells and in the bone environment to inhibit breast cancer growth in the bone. We took advantage of an in-vivo model of bone metastases with SCP1833 subclone of MDA-MB231 breast cancer cells. This is an aggressive breast cancer model which after intra-cardiac injection has a high propensity to home and proliferate in the BM. Cancer cells were engineered to express luciferase and GFP and metastatic dissemination and proliferation was followed by in-vivo bioluminescence.
We used wide-genome approaches to investigate how breast cancer cells affect BM stoma when they home to the endosteal niche and how bone microenvironment can influence tumor cells. To analyze the effect of BM environment on disseminated tumor cells, we compared the transcriptional profile of MDA-MB231 SCP1833 from the mammary fad pad, to mimic the primary tumor, with tumor cells isolated from the BM. Importantly, thanks to GFP expression, tumor cells were directly isolated from the specific environment to be representative of the in-vivo situation. Among the different transcripts up-regulated in BM seeding tumors we focused on ID1 and ID3 as potential targets. The IDs are a family of transcription factor inhibitors that has been described to correlate with poor prognosis in breast cancer and have a role in lung metastases formation. We observed that both ID1 and ID3 are up-regulated in MDA-MB231 SCP1833 compared to the parental cell lines. Moreover, the knock-down of ID1 expression in tumor cells is sufficient to decrease the metastatic potential of MDA-MB231 1833 in-vivo. We are following up these results and we will investigate the molecular mechanism involved in ID1 and ID3 expression as well as their impact on bone marrow environment and metastatic progression.
Concomitantly we are interested in investigating potential pathways modulated in the BM environment after seeding and proliferation of tumor cells. For this purpose we compared the transcriptional profile of BM stroma from the bone of tumor bearing vs tumor free mice. Different BM cell populations were isolated by cell sorting thanks to specific markers. We excluded cells of the hematopoietic compartment from the analysis, concentrating on the CD45- population. We used the best available antibodies that define osteoblasts (CD45-, TR119- Sca1-, CD51+), endothelial cells (CD45-, TR119-, Sca1+, CD31+) and mesenchymal progenitors (CD45-, TR119-, CD31-, Sca-1+). Microarray analysis revealed that BM is strongly affected by breast cancer cell dissemination. Indeed, several components of key molecular pathways involved in tumor development (including TGFBR, PDGFRB, EGFR, HGFR and JAK-STAT signaling pathways) are modulated in the BM environment of mice with tumor cells. We are following up this analysis by testing different small molecule inhibitors to block specific pathways in bone marrow metastases.