PhD Candidate MIT Cambridge, Massachusetts, United States
Introduction: The success of checkpoint blockade has revolutionized cancer therapeutics, underscoring the power of immunotherapy. However, this success is limited to patients with “hot” tumors, i.e., tumors characterized by T cell infiltration prior to treatment. In contrast, patients with “cold” tumors, i.e., tumors which lack T cell infiltration, do not respond to this treatment modality. There is thus a need to develop therapies that convert “cold” tumors to “hot” ones. Given the prerequisite to T cell infiltration and activation is myeloid cell activation, developing methods to boost macrophage and dendritic cell activation has become an area of intense study. The receptor tyrosine kinase family comprised of the TYRO3, AXL, and MERTK receptors (TAMRs) is an attractive target for activating macrophages and dendritic cells because the TAMRs are known to be the innate immune system counterparts to PD-1 and CTLA-4 of the adaptive immune system. That is, the TAMRs are innate immune checkpoints. Furthermore, TAMRs are known to promote tumor cell survival, invasion, metastasis, drug resistance, and immune evasion. Thus, TAMRs are dual targets and many therapies targeting these receptors are in clinical trials. However, the effects of TAMR inhibition on the tumor microenvironment, and more specifically on the immune compartment, are incompletely understood. In this work, we developed a human in vitro model system to elucidate the effects of TAMR inhibition on tumor cells, macrophages, dendritic cells, and the crosstalk between these cell types.
Materials and
Methods: A375 tumor cells were incubated for 24 hours prior to collection of cell lysates which were assayed for TAMR quantification via ELISA. Primary human monocyte-derived macrophages (HMDMs) were generated by culturing CD14+ monocytes with 25 ng/mL M-CSF for 6 days. HMDMs were polarized to a pro-inflammatory state (M1) with 10 ng/mL lipopolysaccharide (LPS) + 20 ng/mL interferon-gamma (IFN-g), to an anti-inflammatory state (M2) with 20 ng/mL interleukin-4 (IL-4) + 20 ng/mL IL-13, or to a tumor-conditioned state with A375 conditioned media for 24 hours. The cells were then lysed and assessed for TAMR expression via ELISA. Untreated A375 tumor cells were cultured in DMSO vehicle for 24 hours followed by 24 hours rest. Injured A375 tumor cells were cultured in 15 nM Trametinib + 150 nM Dabrafenib for 24 hours followed by 24 hours rest. Dead A375 tumor cells were cultured in 15 nM Trametinib + 150 nM Dabrafenib for 72 hours. Prior to co-culture, A375 tumor cells were labeled with pHrodo Red, a pH-sensitive dye. Cells were treated with DMSO vehicle, Bemcentinib, an anti-AXL inhibitor at 1 uM, or BMS-777607, an anti-TAMR inhibitor at 10 uM. Phagocytosis was then assessed via flow cytometry, and HMDMs were also assessed for changes in various surface marker expression. Primary human monocyte-derived dendritic cells (HMDDCs) were generated by culturing CD14+ monocytes with 40 ng/mL GM-CSF + 20 ng/mL IL-4 for 8 days. Following tri-culture, HMDMs and HMDDCs were assayed for changes in various surface marker expression via flow cytometry.
Results, Conclusions, and Discussions: We selected the A375 melanoma cell line for our model because A375 tumor cells overexpress AXL (Figure 1A). HMDMs express both AXL (Figure 1B) and MERTK (Figure 1C), and this expression is modulated by polarization state. Furthermore, both receptors are expressed when HMDMs are cultured with conditioned media from untreated, injured, and dead A375 tumor cells, underscoring the need to investigate effects of TAMR inhibition in both the tumor and immune compartments of the tumor microenvironment. TYRO3 was consistently below the limit of detection. Given our ultimate interest in studying the effects of TAMR inhibition in tumor-immune crosstalk, we co-cultured the two cell types. We were first interested in how TAMR inhibition may affect HMDM consumption of A375 tumor cells. Interestingly, while both inhibitors appear to yield a decrease in untreated A375 consumption (Figure 1D), Bemcentinib results in an increase in consumption of injured (Figure 1E) and dead (Figure 1F) tumor cells. We were also able to assess HMDM surface marker expression and observed that treatment with Bemcentinib but not BMS-777607 yields slight changes in expression as compared to DMSO control, which can be analyzed and visualized multivariately via principal component analysis (Figures 1G-I). However, there is not a clear shift towards a pro- or an anti-inflammatory state, highlighting the need for transcriptomic studies to better characterize nuances in HMDM state. Finally, we have begun conducting tri-culture experiments with HMDDCs, and preliminary data shows that this additional cell type appears to drastically alter the system. While in co-culture treatment with untreated A375 tumor cells yields increases (fold change >1) in HMDM CD206 and CD40 expression, in tri-culture HMDM expression of these markers decreases (Figures 1J-K). Furthermore, Bemcentinib appears to affect HMDDC state (Figures 1L-O) and may influence HMDDC maturation. Clearly, the system is complex and while in co-culture AXL inhibition may result in a pro-inflammatory, anti-tumor response, in tri-culture these beneficial effects may be subdued. We are currently working towards understanding how TAMR inhibition affects the cytokine milieu by analyzing cell secretomes via Luminex, and we will assess how TAMR inhibition alters cell-cell communication via transcriptomics.
