Abstract
NK cell cytotoxicity is mediated by release of granules containing perforin and granzymes and/or expression of death ligands FasL and tumor necrosis factor-related apoptosis-inducing ligand (TRAIL). TRAIL expression is considered desirable on expanded NK cells, although a direct contribution to cytotoxicity will vary with target cell expression of death-inducing TRAIL receptors (DR4 and DR5) versus regulatory, decoy receptors (DcR1, DcR2 and OPG), and sensitivity to rapid lytic granule-mediated killing (Prager et al. 2019). Moreover, a limited reservoir of pre-formed TRAIL may restrict its contribution to overall killing. NK cell-expressed TRAIL has also recently been implicated in enhancing degranulation and cytokine release, independent of a canonical role in triggering target cell apoptosis (Höfle et al. 2022).
We set out to enhance NK cell activity in acute myeloid leukemia by ectopic expression of wild-type TRAIL (WTT) or a potent high-affinity DR4 specific TRAIL variant (DR4v, 4C9) in peripheral (PB)- and umbilical cord blood (CB)-derived NK cells (Reis et al. 2010).
Purified PB NK cells or CB mononuclear cells were activated and expanded by co-culture with EBV-LCL feeder cells and IL-2. To establish the importance of TRAIL to baseline NK cell cytotoxicity, we applied concanamycin A to inhibit perforin and granule mediated killing and knocked out the only surface-expressed TRAIL death-inducing receptor present on U937 AML cells (DR5). Relative to baseline cytotoxicity (82%), ConA (22%, p=0.001) led to a greater reduction in U937 killing than DR5 KO (62%, p=0.05), consistent with granule release being the dominant mechanism of cytotoxicity (E:T-1:1, 20 h, n=4). We then evaluated WTT overexpression in the NK cell line KHYG-1 using mRNA electroporation, achieving a 5.7-fold increase in TRAIL expression (4 h post-transfection) and relative increases in cytotoxicity ranging between 13% and 62% in KHYG-1 sensitive AML cell lines (n=3). KHYG-1-resistant cell lines remained resistant.
To evaluate ectopic WTT or DR4v expression in a clinically relevant platform, we applied the TcBuster™ transposon system to express a bicistronic construct comprising of a mutant dihydrofolate reductase (DHFR) and either WTT, DR4v, or low affinity growth factor receptor (negative control) separated by a P2A sequence in primary NK cells. Transposed cells expressing DHFR were selected for by methotrexate during feeder stimulation. Neither TRAIL construct led to loss of NK cell viability versus control. Ectopic expression of WTT (2.0-fold, p=0.04) or DR4v (2.1-fold, p=0.04) in CB NK cells increased cell-surface TRAIL relative to control, with further increases after treatment with the matrix metalloprotease inhibitor TAPI-1- suggesting shedding limits surface TRAIL accumulation (panel A). Expression of DR4v enhanced CB NK cell cytotoxicity against EM2, Kasumi-1 and THP-1 AML cell lines (panel B). Using primary AML samples from patient donors, increased cytotoxicity was detected for DR4v TRAIL (specific cytotoxicity 27% vs 13%, E:T of 1:1, p=0.003), and a trend for WTT overexpression (19%, p=0.06) (panel B). In PB NK cells, a similar pattern of TRAIL expression and enhanced primary AML blast-specific cytotoxicity was observed (control 32% vs. DR4v 43%, p=0.01).
We hypothesized that these differences may be partly attributable to NK cell intrinsic, non-canonical effects of increased TRAIL expression. Accordingly, we noted that TRAIL-engineered NK cells express significantly higher levels of NKp30 and CD16 with a trend for increased NKG2D expression (panel A). After 4-hour co-culture with AML targets, both WTT- (24.2%, p=0.09) and DR4v-expressing (25.1%, p=0.03) primary NK cells showed enhanced degranulation versus control NK cells (16.4% CD107a+ NK cells, n=4). CD107a+ subpopulations for control, WTT- and DR4v-expressing primary NK cells consistently displayed higher cell-surface TRAIL relative to CD107a- populations.
In summary, NK cell anti-leukemic activity was enhanced through TRAIL engineering, primarily through expression of a potent high-affinity DR4 receptor specific TRAIL variant, which significantly enhanced innate, antigen-independent elimination of primary AML cells. Our data support a role for TRAIL and TRAIL variants in enhancing NK cell activation and degranulation independent of death receptor signaling and suggest a future role for TRAIL engineered NK cell therapies.
Disclosures
O'Reilly:ONK Therapeutics: Current Employment. Brophy:ONK Therapeutics: Current Employment. Hermanson:Bio-Techne: Current Employment. Szegezdi:ONK Therapeutics: Research Funding. O'Dwyer:ONK Therapeutics: Current Employment, Current equity holder in private company, Membership on an entity's Board of Directors or advisory committees.
