Genetic modification of human natural killer cells and possible applications thereof

Sammanfattning: Natural killer (NK) cells are dependent on the presence of interleukin-2 (IL-2) for their survival and for their cytotoxicity against tumor cells. Currently, for in vitro expansion of NK cells from peripheral blood mononuclear cells (PBMCs), IL-2 is of great importance. Moreover, during the infusion of NK cells back into patients, systemic IL-2 administration enhances their proliferation and immunological properties. However generic IL-2 administration is frequently impeded by undesirable side effects, such as high toxicity and undesired activation of neighboring cell populations. The latter is extremely unfortunate especially after allogeneic hematopoietic stem cell transplantation, where activation of, e.g., T cells and Treg cells could lead to GVHD and immunosuppresion. The aim of this thesis was to investigate whether NK cell activity can be improved by transduction with IL-2 with regard to in vitro expansion leading to the independence from exogenous IL-2 administration and thus a more promising cancer therapy solution. However, since secreted interleukins have proven to be non-specific, leading to undesirable toxic side effects, we reasoned it would be better if interleukins were expressed in a localized and controlled manner. Initially, as proof of principle, the IL-2-dependent NK-92 cell line was transduced with wild type IL-2 (IL-2WT) or an IL-2 vector targeted to subcellular organelles, including ER (IL-2ER). NK-92 cells were also transduced with GFP as a positive control. When populations of modified and non-IL-2 modified NK-92 cells were mixed, it was shown that the ER-targeted IL-2 did not support the growth of neighboring NK-92 non-IL-2-modified cells. This indicated that, in the case of endoplasmic reticulum targeting, the IL-2 does not leak to the supernatant. Next, in order to establish an efficient protocol for retroviral transduction of primary NK cells, PBMCs, including NK cells were transduced with GFP on several timepoints during a 21day expansion, with day 5 showing the best transduction rate. In order to test the above IL-2 constructs in vivo, an animal model had to be established, where cells could be monitored. For this purpose 5T33MM cells were transduced with GFP, injected into C57B1/KaLwRij mice, and monitored for homing and mice survival purposes. Finally this GFP MM mouse model was used to assess the therapeutic potential of IL-2-induced NK cells. NK cells separated from single-cell suspensions derived from mouse organs, and activated with IL-2, were shown to eliminate syngencic myeloma cells in vitro and to prolong the survival of 5T33MMGFP +-immunized mice in vivo.