Abstract
Macrophage Migration Inhibitory Factor (MIF) is an inflammatory cytokine expressed by most cell types, which functions as both a cell signaling molecule and a chemokine. MIF is known to have an intrinsic enzymatic activity as a keto-enol tautomerase, however the substrate for this activity in vivo is unknown, and the importance of this activity in the biological function of MIF is still debated. Overexpression of MIF has been found in both the serum and primary tumor tissue of patients with numerous cancer types, including lung, ovarian, colorectal and breast cancer, and increased expression often correlates with poor outcomes and increased risk of metastasis. Increased MIF expression has also been shown to drive tumor progression and enhance metastasis in a variety of mouse models of cancer, further supporting a role for MIF in tumor progression. Many of the studies linking increased MIF expression to enhanced tumor progression have established MIF as an immune modulator, effectively working to dampen the anti-tumor immune response through a variety of mechanisms. Our group has demonstrated previously that MIF expression in the primary tumor leads to enhanced accumulation of the monocytic subset of immunosuppressive myeloid derived suppressor cells (MDSCs) in the primary tumor, leading to decreased T cell activation, and increased tumor growth. We have now discovered that loss of MIF expression in the 4T1 cell line also promotes the cancer cells to undergo a specific type of cell stress-induced death, termed “immunogenic cell death” (ICD). We show that loss of MIF expression enhances the ICD response specifically when cells are cultured under serum-free conditions in vitro. Furthermore, we demonstrate that when implanted in vivo, MIF-depleted 4T1 tumors show enhanced infiltration of activated dendritic cells and T cells, indicative of an enhanced anti-tumor immune response. We suggest a model whereby MIF expression in the primary tumor protects the cancer cells from undergoing ICD in vivo, resulting in a dampened anti-tumor immune response, and loss of control of tumor growth. Numerous studies, both in human cancer patients and in mouse models of cancer, also support a role for MIF in promoting metastasis. Supplementary to this, we have previously reported that depletion of MIF in the 4T1 model results in an almost complete loss of spontaneous pulmonary metastasis. However, we have since discovered that when MIF depleted tumors are permitted to grow to the same (larger) size as a MIF-expressing tumor, pulmonary metastasis is restored to the levels observed in the setting of a MIF-expressing tumor. This suggests that it is tumor size, rather than MIF expression, that is driving metastasis. In these same studies, we also analyzed whether MIF expression in the primary tumor had any effect on the generation of a metastatic niche, a process by which the primary tumor prepares a distant organ for future metastasis. Using the 4T1 model, we discovered that while loss of MIF expression in the primary tumor has no effect on collagen matrix remodeling in the lungs, loss of MIF expression does lead to a reduction in overall accumulation of CD11b+ myeloid cells in the lungs, both early and late during tumor development. Importantly, we found that the decrease in myeloid cells observed is not dependent on primary tumor size, suggesting that MIF expression in the tumor is directly driving this phenotype. Further analysis of the myeloid cell populations in the lungs of WT versus MIF KD tumor-bearing mice by flow cytometry revealed no differences in the subsets analyzed. More detailed analysis of lung-infiltrating myeloid cells will need to be performed to determine which subset(s) is specifically controlled by MIF expression in the primary tumor. We have also utilized the MMTV-PyMT murine transgenic model of breast cancer to further confirm the role of MIF in tumor progression. We observed a significant delay in tumor progression in MIF KO mice compared to WT mice using this model. We also detected a significant decrease in overall tumor burden at both 8 weeks and 5 months in MIF KO mice. However, we discovered no difference in pulmonary metastasis between WT and MIF KO mice at the late stages of tumor development, suggesting that MIF expression is important for promoting primary tumor growth, but not metastasis. Based on our previously published work, we hypothesized that MIF expression may promote tumor growth through increased accumulation of MDSCs in the tumor microenvironment in this model. However, when we analyzed tumor-infiltrating myeloid cells both early and late during tumor development, we found no significant differences in either MDSC subpopulation. This suggests that there is another mechanism by which MIF is controlling tumor growth in this model. We propose that a more detailed analysis of immune infiltrates in early stage tumors in this model, focusing on dendritic cells and macrophages, will begin to reveal the mechanism by which is MIF is promoting growth. Increased dendritic and T cell activation may also suggest that MIF expression in this model is functioning to protect cancer cells from undergoing ICD, as we see in the 4T1 model, and would further support a model in which loss of MIF expression renders the primary tumor more susceptible to recognition and attack by the immune system. Overall, our work strongly supports development of a clinical MIF inhibitor for use in the setting of treating patients with solid tumors. We show a clear role for MIF in primary tumor growth, which is reliant on interaction with the host immune response. Immunotherapy has proven to be quite successful in treating a number of cancer types in the clinic; however, the immunosuppressive tumor microenvironment remains a challenge to overcome in patients who do not show an ideal response to these treatment modalities. Our work suggests that combination of MIF inhibition with current immunotherapeutic strategies could enhance responses, and lead to better outcomes for a large number of patients.
