(D-136) Collagen Mineralization Inhibits the Capacity of Macrophages to Stimulate Breast Cancer Cell Invasion

Introduction: As of 2021, breast cancer is the most common cancer worldwide. While primary breast cancer can be treated relatively effectively, the 5-year survival rate of patients with metastatic disease is less than 30%, with most of these patients developing skeletal metastases. To colonize bone, disseminated tumor cells extravasate from the vessel and invade through the bone marrow to seed osteogenic niches. Tumor cell invasion during this process is controlled by host cells including macrophages. Macrophages are particularly relevant as they not only exert immunomodulatory effects, but also have the potential to differentiate into osteoclasts, the primary cell type driving osteolysis in bone metastasis patients. Importantly, reduced bone mineral density is a risk factor for breast cancer bone metastasis and previous work has demonstrated that macrophages interact with mineralized bone matrix. However, how changes in bone matrix mineralization affect the ability of macrophages to regulate breast cancer invasion remains largely unclear due in part to a lack of relevant experimental models. In this study, a microfluidic model system was established in which mineralization of collagen, the primary component of bone matrix, as well as tumor cell and macrophage seeding could be selectively controlled. Utilizing this model system, the ability of macrophages to regulate tumor cell invasion as a function of bone matrix mineralization was investigated.

Materials and Methods: A polydimethylsiloxane (PDMS) mold was fabricated using soft lithography. Subsequently, collagen type I was cast into the central region of the device to form two separate microchannels using a needle pull-through technique. Collagen mineralization in the right channel of the device was induced using a polymer-induced liquid precursor (PILP) method [1,2] that results in physiologically relevant, intrafibrillar collagen mineralization (Figure 1A). Mineralization was confirmed using alizarin red staining in combination with scanning electron microscopy (SEM). Human MDA-MB-231 breast cancer cells were labeled with CellTracker488 and seeded into the left, non-mineralized collagen channel. The right channel of the device was either mineralized or not and then seeded with or without murine RAW264.7 macrophage cells. After culturing for 5 days, devices were fixed, stained with DAPI and F4/80, and imaged using confocal microscopy. Image analysis was conducted using ImageJ.

Results: Alizarin red staining (Figure 1B) and SEM (not shown) confirmed that the PILP method allowed for selective mineralization of one channel of the device only. Confocal reflectance verified that the lumen structure of the mineralized channel remained unchanged (Figure 1C). Seeding of MDA-MB-231 cells or RAW264.7 cells into monoculture devices confirmed that each cell type remained confined to one channel. Co-culture experiments subsequently identified that RAW264.7 cells promote MDA-MB-231 invasion regardless of whether they were seeded into a non-mineralized or mineralized channel (Figure 1D). However, this effect was less pronounced when RAW264.7 cells were seeded into a channel that was previously mineralized. The presence of mineral did not affect MDA-MB-231 invasion when cultured without RAW264.7 cells (Figure 1E).

Discussion: The data presented here demonstrate that microfluidic collagen devices can be fabricated with or without selective mineralization of one channel to study how changes in bone mineral matrix content affect tumor-host cell interactions. Furthermore, we confirmed that macrophages promoted the invasion of breast cancer cells regardless of collagen mineralization, consistent with findings by others. This effect occurred without cell-cell contact, suggesting that it was caused by soluble factors secreted from macrophages. Additionally, it was determined that collagen mineralization inhibits the ability of macrophages to promote tumor cell invasion. These findings suggest reduced bone mineral density may increase risk of bone metastasis by promoting tumor invasion within the bone marrow niche. While the current study used a cell line, we are currently repeating experiments with primary murine bone marrow-derived macrophages and syngeneic breast cancer cells to validate the relevance of our findings.

Conclusions: The research presented here describes a new model system that recapitulates key microenvironmental features of the bone metastatic niche. This model system is highly adaptable and can be modified by introducing additional cell types and/or soluble factors to probe other aspects of bone and tumor cell interactions. Such studies have the potential to yield novel insights that will inform therapeutic strategies to target disseminated tumor cells during early stages of metastasis.

References:
1. Choi, S. et al. Intrafibrillar, bone-mimetic collagen mineralization regulates breast cancer cell adhesion and migration. Biomaterials. 198, 95-106 (2019).
2. Choi, S. et al. Biofunctional matrix models reveal mineral-dependent mechanoregulation of bone metastatic breast cancer. bioRxiv (2023).