Natural Killer (NK) cells have gained much recent attention as a promising alternative platform for CAR engineering owing to their unique biological attributes, their specialized cytotoxicity against tumors, their safety profile, and their potential use as an off-the-shelf cellular therapy. However, NK transgene expression level is a major technical obstacle to the development of NK cell immunotherapy. Promoters, present in vectors, are responsible for maintaining the expression of CAR in host cells. Furthermore, promoters may differ in their ability to produce lentiviral particles and in their transduction efficiency. The present study aimed to evaluate whether different internal promoters result in a different percentage of transduced cells and transgene expression level in NK cells and thus to be able to choose an ideal promoter to drive the expression of CAR.CD19 increasing the antitumor potency of NK cells.

Lentiviral particles with EF1a and SFFV promoters driving GFP gene expression or CAR.CD19 were produced, and the virus titer was measured by flow cytometry. Lentiviral particles were used for transduction of NK-92, NK from peripheral blood and NK from cord blood. CAR positive cells were detected by flow cytometry using anti F(ab)2 antibody. CAR-NK cell cytotoxic potential and specificity were assessed using established human Raji and NALM-6 lymphoblastoid cell lines (CD19+) as targets, which express varying antigen levels and MDA-MB-453 (CD19-) cells were used as negative control.

First, we explored the potential EF1a and SFFV promoters to drive GFP gene expression. SFFV promoter produced high lentiviral titers and achieved the best transduction efficacy in NK cells. The major difference between the promoters was the ability of SFFV to maintain longer and more robust GFP expression in NK cells. Therefore, we chose this promoter to drive CAR.CD19 expression and interleukin (IL)-15 co-expression in NK cells. There was a transduction efficiency of 13% (+/- 13.7) for NK-92 cells, assessed 48h after transduction. NK cells from Peripheral Blood (NK-PB) and Cord Blood (NK-CB) were also transduced with an efficiency of 27% and 49%, respectively. After 3 weeks in culture, the percentage of CAR-NK cells progressively decreased in primary cells. Cell growth of transduced cells from both sources was not statistically different; despite wild type NK-CB cells had a higher fold-change expansion than wild type NK-PB. CAR-NK-92 cells modified with SFFV were enriched with magnetic beads to 98% of NK cells positive for CAR expression and showed efficient and stable transgene expression during ex vivo expansion. The enriched CAR.CD19-NK-92 cells showed a higher killing potential against Raji CD19+ cells, compared with non-transduced NK-92 cells. Further, we evaluated the CAR-NK primary cells in vitro cytotoxic potential. CAR-NKs both from PB and CB had a higher killing potential against Raji and NALM-6 CD19+ cell lines than non-transduced NK cells.

Taken together, these results indicate that the developed vector is functional and able to transduce NK cells from different sources, enhancing their anti-tumoral activity against tumoral CD19+ cells. Our results highlight that the promoter choice is essential to generate CAR-NK cells suitable for therapeutic applications.

Supported by FAPESP 2019/25309-0, 2013/08135-2, 2008/578773; CNPq 573754-2008-0; Capes (88887.140966/2017-00 and 88881.199630/2018-01).