Optimizing Lentiviral Transduction of Human Natural Killer Cells

Abstract Abstract 4714 The development of an efficient method to genetically modify natural killer (NK) cells could be used to characterize NK cell differentiation, acquisition of self-tolerance, tumor trafficking in vivo, as well as to manipulate NK cells to enhance their activity against infectious diseases and tumors. Although HIV-1 based lentiviral vectors (LVs) have been used to efficiently transfer genes into human T-cells, little data exists on LV transduction of either fresh or in vitro expanded human NK cells or its effects on NK cell phenotype and cytolytic function. In this study, we used an HIV-based LV expressing enhanced green fluorescence protein (EGFP) driven by a murine stem cell virus long terminal repeat (MSCV-LTR) promoter to transduce CD3− and CD56+ and/or CD16+ human NK cells that were either resting, IL-2 activated, or expanded in vitro using an irradiated EBV-LCL feeder cell line. We observed that resting NK cells were difficult to transduce with LVs, even at high multiplicities of infection (MOI), with transduction efficiencies (TE) in the range of only 3–14%. The efficiency of LV transduction improved when the NK cells were pre-stimulated in vitro with IL-2: TE improved to 21±0.2% in NK cells cultured for 24 hours in media containing IL-2 (200 U/mL) and 28.7±12.9% in NK cells that underwent in vitro expansion over 9 days prior to transduction using irradiated EBV-LCL feeder cells and media containing IL-2 (200U/mL). Subsequently, we evaluated incremental MOIs (3-200) to optimize LV transduction of expanded NK cells; optimal transduction was achieved using a spinoculation protocol at a MOI of 25 which resulted in the highest transduction efficiencies with the least amount of cell death. Increasing the MOI above this level resulted in a small increase in transduction, but was offset by an increase in NK cell apoptosis/death. Using a one-round, non-spinoculation protocol and an MOI of 30, we obtained a median transduction efficiency of 29% (range 16–41) with excellent retention of NK cell viability. This optimized protocol was used to transduce expanded NK cells with a LV vector encoding an shRNA targeting a region of the NK cell inhibitory receptor transcript NKG2A. Following transduction, surface expression of NKG2A decreased significantly on expanded NK cells compared to non-transduced expanded NK cells and “scramble transduced” LV controls; at a MOI of 10, the MFI of NKG2A on expanded human NK cells decreased 35% compared to non-transduced and LV transduced scramble controls (median MFI 428, 673, 659 in shRNA, non-transduced and scramble LV control transduced NK cells respectively). A comparison of transduction efficiencies using LVs expressing EGFP driven by MSCV-LTR, EF1a, and Ubi promoters showed MSCV-LTR mediated the highest level of gene expression in expanded NK cells. Transduced NK cells maintained stable EGFP transgene expression in vitro, which peaked 5 days following LV transduction and remained stable for an additional 9 days. The phenotype of lentiviral transduced NK cells was similar to non-transduced NK cells. Specifically, expression of CD56, CD16, granzyme A and B, perforin, the inhibitory receptors NKG2A, KIR3DL1, KIR3DL2, and KIR2DL1/DL2, and the activating receptors NKG2D, NCRs NKp46, and NKp30 were not altered in either fresh or expanded NK cells following LV transduction, although we did observe a significant reduction in NKp44 expression in LV transduced cells (22% compared to 50% on untransduced NK cells; 0.02). Furthermore, NK cell function, as assessed by cytokine production and cytotoxicity vs tumor targets was not altered in LV transduced NK cells. A 51Cr release cytotoxicity assay showed GFP+ NK cells, flow sorted following LV transduction of expanded NK cells, had similar cytotoxicity against K562 cells and human renal cell carcinoma cells (RCC) compared to non-transduced expanded NK cell controls (figures). In conclusion, we show that an HIV-1 based lentiviral vector driven by a MSCV-LTR, mediated efficient and stable gene transfer in IL-2 activated and in vitro expanded human NK cells. This study provides valuable insights for methods to optimize the long-term expression of LV transduced genes in human NK cells which could be used to improve their anti-tumor function in vivo. Target: K562 cells Target: RCC cell line Disclosures: No relevant conflicts of interest to declare.