Using Advanced Microscopy Techniques for the Study of Macrophage-Cancer Cell Interactions in the Presence of Therapeutic Antibodies

The use of monoclonal antibodies represents a rapidly expanding area for cancer therapy. One of the main mechanisms of action of these antibodies is Fcγ receptor-mediated engagement of macrophages and other immune cells. When macrophages engage tumor cells opsonized with antibody molecules, they can perform trogocytosis, the process of internalizing fragments of the target cell, or phagocytosis, the internalization of entire cancer cells. This study first establishes whether the process of trogocytosis can lead to cancer cell death. A variety of microscopy and flow cytometric assays were used to quantify the levels of trogocytosis and cell death, in co-cultures of macrophages and cancer cells. Using HER2-overexpressing breast cancer cell lines and anti-HER2 antibodies, we show that persistent trogocytosis can lead to the killing of cancer cells. The mechanism of trogocytosis was also explored using multifocal plane microscopy (MUM). Imaging the process of trogocytosis using MUM revealed that it proceeds through the macrophage-mediated extrusion of tubular structures of the target cell membrane. This membrane-tubulation results in the preferential uptake of the membrane components from the target cell. The study also investigated the maturation pathway followed by phagosomes containing entire cancer cells. A vacuole-like structure associates with these phagosomes, which whilst also lysosomal in nature, displays characteristics distinct from the phagosome itself. The interface between the vacuole and the phagosome is impermeable to certain solutes as observed through microscopy. Further, the size of the phagosome-associated vacuole is affected by inhibition of the mTOR pathway. Use of advanced microscopy techniques such as MUM in these and other biological problems provides mechanistic insight at the spatiotemporal level. To further develop the algorithms involved in MUM data processing, I have therefore also explored various non-parametric methods of estimating the axial location of point sources from MUM data. A new non-parametric method is proposed, which uses multiple intensities calculated from each image of a point source in MUM data. The performance of this approach is compared with other non-parametric methods through simulations and Fisher information calculations. The effectiveness of this method on experimental data is also evaluated.