Targeting the Interferon-γ (IFN-γ) Pathway to Enhance Susceptibility of Multiple Myeloma Tumor Cells to NK Cell Lysis

Abstract 4013

Multiple myeloma (MM) is a B cell neoplasm characterized by clonal expansion of malignant plasma cells in the bone marrow. Despite the use of effective new agents, MM remains an incurable disease. NK cells are an important element of immune surveillance, capable of killing virus-infected and malignant-transformed cells. Understanding of how target cells respond to NK cells will help improve clinical outcome in cancer patients and lead to further advances in the clinical application of NK cell therapy to prevent and treat MM. To identify novel pathways that modulate MM cell resistance to human NK cells, we previously screened a large lentiviral shRNA library to identify genes that play a role in this cell-cell interaction. In this genetic screen, silencing JAK1 and JAK2 in MM target cells significantly increased their susceptibility to NK cell lysis. JAK1 and JAK2 are members of a family of tyrosine kinases that are constitutively associated with membrane cytokine receptors and become active only when the cell surface receptor is activated by a specific ligand. In these conditions, activated JAKs induce phosphorylation of STAT proteins, which subsequently initiate gene transcription programs.

To determine how JAK1 and JAK2 modulate MM cell resistance to NK cells, we undertook experiments to analyze the JAK signaling pathway in MM cells. We first analyzed the activation status of STAT proteins in a series of MM cell lines (IM-9, KM12BM, RPMI 8226, MM1S) in which JAK1 and JAK2 expression was reduced by specific shRNAs. Constitutive activation of STAT proteins was not affected by JAK1 or JAK2 silencing suggesting that these kinases were not activated in the absence of cytokine receptor-mediated signaling. Since JAK1 and JAK2 are associated with the IFN-γ receptor and NK cells are known to secrete IFN-γ after activation by target cells, we hypothesized that these kinases were specifically activated when NK effector cells became engaged with MM target cells. To test this hypothesis, MM target cells were stimulated with NK-activated supernatant or recombinant IFN-γ and examined for STAT, AKT and ERK activation. Incubation with either NK-supernatant or IFN-γ induced strong phosphorylation of STAT1 but other STATs, AKT or ERK were not activated. Silencing of JAK1 or JAK2 with specific shRNAs prevented STAT1 activation. These findings were validated in primary MM cells treated with Jak inhibitor 1. When compared with MM cell lines, primary MM cells had a similar STAT profile at their basal level and incubation with IFN-γ or NK-activated supernatant only induced rapid activation of STAT1, which was inhibited when MM cells were pre-treated with Jak inhibitor 1. Treatment of primary MM cells with Jak inhibitor 1 also increased NK cell-mediated killing. To confirm that IFN-γ secreted by activated NK cells induced resistance in MM target cells, we tested the ability of anti-IFN-γ antibody (D9D10) to block STAT1 phosphorylation in MM cells pre-incubated with IFN-γ or NK-activated supernatant. In these experiments, D9D10 completely blocked STAT1 phosphorylation when compared to an anti-IgG1 isotype control. We then co-cultured MM cells with primary NK cells and D9D10 or IgG1 isotype control antibody and evaluated target cell killing by AnnexinV/7AAD staining. In experiments with IM-9 and KM12BM target cells we found that 2 and 5 μg of IFN-γ blocking antibody increased specific target cells lysis by 37.6%, 46.5% and 26.7%, 29.5% respectively compared to IgG1 isotype control. Similarly, addition of 2 and 5 μg of blocking IFN-γ antibody to primary MM cells co-incubated with primary NK cells increased killing by 28.8% and 38.3% compared to controls. These data suggest that IFN-γ enhances resistance of MM cells to NK cell-mediated killing and that inhibiting this pathway at the level of JAK1, JAK2 and possibly STAT1 can reverse this effect resulting in increased killing of MM tumor cells.