Role of Gilt in Melanoma Cells on Regulating in Vivo Tumor Growth

Abstract

The MHC I antigen presentation pathway in melanoma cells has a well-established role in immune-mediated destruction of melanoma. However, the role of the MHC II pathway in melanoma cells is not fully understood. Gamma-interferon-inducible lysosomal thiol reductase (GILT) is critical for MHC II-restricted presentation of multiple melanoma antigens by antigen presenting cells. While GILT and MHC II expression is typically limited to antigen presenting cells, GILT and MHC II can be expressed constitutively or induced by IFN-γ in melanoma cells. In human melanoma specimens, high GILT expression and an active and intact MHC II pathway are associated with improved survival. The goal of this project was to investigate a causal role for GILT in melanoma cells, using immunogenic Yale University Mouse Melanoma (YUMM) lines YUMM2.1, YUMMER1.7, and YUMMER.G. These lines constitutively expressed GILT, and MHC II expression was IFN-γ-inducible in a subset of cells under serum-free conditions. YUMM2.1 was selected as the primary model to assess the role of the MHC II pathway in in vivo tumor growth studies, because YUMM2.1 cells reliably formed tumors in mice. We have genetically engineered YUMM2.1 cells via CRISPR/Cas9 without GILT, verified the deletion, and selected a clone (#6) with similar in vitro proliferation as wild-type (WT) YUMM2.1 cells. When GILT-/- #6 cells were injected into immunocompetent mice, the resulting tumors had increased growth over time compared to WT tumors, but there was no clear association between tumor type and tumor onset or mouse survival. When these cells were injected into immunocompromised RAG1-/- mice, the resulting tumor growth, onset, and mouse survival, were inconsistent between experiments. Although further studies are needed to address these inconsistencies, in both experiments GILT-/- #6 had increased tumor growth in immunocompetent mice compared to WT tumors. Flow cytometry analyses showed that GILT-/- #6 tumors had fewer infiltrating T cells compared to WT, and a higher percentage of the infiltrating T cells were regulatory, supporting a potential role for T cells in regulating the in vivo tumor growth of GILT-/- #6 tumors versus WT tumors. However, to verify that the difference in tumor growth we observed in vivo was solely due to GILT expression in melanoma cells, we tested the in vivo tumor growth of GILT-/- YUMM2.1 cells transduced with empty vector or transduced with GILT. We found that GILT expression did not impact in vitro proliferation nor in vivo growth in immunocompromised mice. However, both tumors with the empty vector or GILT were rejected in immunocompetent mice, likely due to the expression of puromycin resistance from the lentiviral plasmid. Thus, mice genetically engineered to express puromycin resistance will need to be used in future studies and we are also designing lentiviral plasmids lacking puromycin resistance or other immunogenic reporter genes. In this manner, we will determine the effect of GILT expression in melanoma cells on regulating in vivo tumor growth.