Development and Evaluation of Mouse Monoclonal Antibodies Against Human C1q

Abstract

The complement protein C1q plays an important role in breast cancer susceptibility, development, and progression. Mice genetically deficient in C1qA exhibit an eighty-five percent decrease in mammary tumour incidence when administered the chemical carcinogen DMBA. Similarly, in the aggressive MMTV-PyMT model of mouse mammary cancer, C1qA null mutant mice exhibit two thirds of the tumour burden seen in wild type mice and a third of the late-stage carcinoma at eighteen weeks of age. There may also be a role for C1q in human breast cancer as women genetically deficient in C1q have a decreased incidence of breast cancer. C1q plays a key role in macrophage-mediated efferocytic uptake of dying mammary epithelial cells, thereby preventing apoptotic cells from becoming necrotic and releasing pro-inflammatory cytoplasmic components. Persistence of necrotic cells alters the immune response during cancer initiation, as evidenced by an increase in cytotoxic T cells mobilised to the mammary gland of C1q null mutant mice in response to carcinogen DMBA. The over-arching goal of this research project was to generate and characterise an inhibitor of C1q-mediated efferocytosis. A monoclonal antibody was chosen as the C1q inhibitor, to further explore the role of C1q in breast cancer and as a potential first step in development of a new breast cancer therapeutic. Human C1q was inoculated into C1qA null mutant mice to generate an antibody-mediated immune response to C1q. Three fusions of immune splenocytes from these mice yielded a total of 2,776 cultures of which 2,017 contained viable hybridomas. Antibody binding by enzyme-linked immunosorbent assay (ELISA) included both linear epitopes (contiguous amino acids) and conformational epitopes (binding to a three-dimensional structure), making this an ideal screening strategy. Screening of these cultures by ELISA identified eight hybridomas that bound C1q. Of these, four were successfully expanded and cultures established from single cells. Thus, four candidate monoclonal antibodies were generated: BHI1-1G4, BHI1-4D3, BHI3-3F6, and BHI3-8B9. Characterisation of these antibodies was performed to determine the specificity of their binding conditions. Binding of the monoclonal antibodies in assays that involve denaturation of proteins and presentation of linear epitopes was not observed. These assays included western blotting, immunohistochemistry on normal breast sections, and immunocytochemistry on a fixed macrophage cell line. An assay with potential to display conformational epitopes, immunocytochemistry on unfixed macrophages, also did not demonstrate monoclonal binding. The antibodies were also tested for binding to C1q by immunoprecipitation. This assay can detect conformational epitopes, and all four monoclonal antibodies were demonstrated to bind soluble C1q by this method. Combined, these studies suggested that all four candidate monocloncal antibodies bind only to intact native C1q and do not bind to denatured antigen. Identification of a monoclonal antibody that inhibits C1q-mediated efferocytosis required development of a bioassay that quantifies the functional activity of candidate antibodies. An in vitro efferocytosis assay was developed involving fluorescent green-labelled macrophages co-cultured with fluorescent red-labelled MDA-MB-231 breast cancer cells induced to undergo apoptosis by cross-linking the TRAIL receptor 2. Co-localisation of red and green fluorescence as an indicator of efferocytosis was investigated in bone marrow-derived macrophages from C1qA replete and null mice. Reduced efferocytosis was observed in macrophages where C1q was absent and was quantified using ImageJ software involving digital masking of images and Boolean calculation of overlap. Hybridoma supernatants from candidate antibodies BHI1-1G4 and BHI1-4D3 were tested in the assay using the mouse macrophage cell line RAW 264.7 labelled green and dying breast cancer cells labelled red. Commercially available anti-C1q antibody 9A7 and hybridoma supernatant from non-C1q binding cell line BHI3-2C12 were also assessed. The antibodies exhibited variable capacity to affect C1q-mediated phagocytosis however whether a specific candidate monoclonal antibody could effectively inhibit C1q was inconclusive due to a high degree of variability in the timing of apoptotic cell uptake. This research led to the generation of four candidate monoclonal antibodies with potential for further pre-clinical research. Future work should concentrate on purification of the antibodies in order to improve investigation of their inhibitory capacity in C1q-mediated efferocytosis bioassays. Studies in mouse mammary cancer models would also provide valuable data on the potential of these candidate antibodies for downstream clinical applications in breast cancer patients.