Genetically Reengineered K562 Cells Significantly Expand Cord Blood (CB) Natural Killer (NK) Cells Ex-Vivo Expanded with Increased NK Cell Activation and Cytolytic Activity Both In-Vitro and In-Vivo: Potential Use In Adoptive Cellular Immunotherapy (ACI)

NK cells play a role in reducing relapse in hematological malignancy following AlloSCT (Dunbar et al, Haematologica, 2008). NK cell limitations include lack of tumor recognition and/or limited numbers of viable and functional NK cells (Shereck/Cairo et al, Ped Bld Can, 2007). NK ACI provide safe and effective therapy against tumor relapse; yet NK cells are limited to specific cancer types and not all patients demonstrate optimal response (Ruggieri et al. Science, 2002; Ljunggren et al. Nat Rev Immuno, 2007). To circumvent these limitations, methods to expand and activate PBMNCs with genetically engineered K562 cells expressing membrane bound IL-15 and 41BB ligand (K562-mbIL15-41BBL [modK562]; Imai/Campana et al, Blood, 2005) have shown to significantly increase NK cells in number and maintain heterogeneous KIR expression (Fusaki/Campana et al BJH, 2009). We have shown that CB NK cells can be activated/expanded and exhibit enhanced cytolytic activity when cultured in a cytokines/antibody cocktail (Ayello/Cairo et al, BBMT, 2006; Exp Heme, 2009).

To evaluate CBNK expansion, activation, cytolytic mechanism and function against Burkitt lymphoma (BL) tumor target and its influence on NK cell mediated in-vitro and in-vivo cytotoxicity in NOD-SCID mice following stimulation with modK562 cells (generously supplied by D.Campana, St Jude's Children's Hospital, Memphis, Tx).

Following 100GY irradiation, modK562cells were incubated 1:1 with CBMNCs in RPMI+IL-2 (10IU/ml) for 7 days in 5%CO₂, 37°C. NK activation marker (LAMP-1), perforin and granzyme B were determined by flow cytometry. Cytotoxicty was determined via europium assay at 20:1 E:T ratio with Ramos (BL) tumor targets (ATCC). The mammalian expression construct (ffLucZeo-pcDNA (generously supplied by L.Cooper, MD, PhD) was transfected to BL cells using lipofectin and selected by zeocin for stable transfection. Six week old NOD-SCID mice received 5×10⁶ BL cells subcutaneously. Upon engraftment, xenografted NOD-SCID mice were divided in 5 groups: injected with PBS (control), BL only, 5×10⁶ wildtype (WT) K562 expanded (E) CBNK cells, modK562 expanded (E) CB NK cells (5×10⁶) and modK562 expanded (E) CBNK cells (5×10⁷). Ex-vivo ECBNK cells were injected weekly for 5 weeks and xenografted NOD-SCID mice were monitored by volumetric measurement of tumor size (Tomayko/Reynolds, Can Chemother Pharmac, 1989), bioluminescent imaging (Inoue et al Exp Heme, 2007) and survival. The survival distribution for each group was estimated using the Fisher exact test.

On Day 0, NK cells (CD56⁺/3^-) population was 3.9±1.3%. After 7 days, modK562 expanded CBNK cells was significantly increased compared to WTK562 and media alone (72±3.9 vs 43±5.9 vs 9±2.4%, p<0.01). This represented a 35-fold or 3374±385% increase of the input NK cell number. This was significantly increased compared to WTK562 (1771±300%, p<0.05). ModK562 ECBNK cells demonstrated increased perforin and granzyme B expression compared to WTK562 (42±1.5 vs 15±0.5%,p<0.001; 22±0.5 vs 11±0.3%,p<0.001, respectively). Cytotoxicity was against BL tumor targets was significantly increased (42±3 vs 18±2%,p<0.01), along with NK activation marker expression, CD107a (p<0.05). At 5 weeks, in-vivo studies demonstrated increased survival of NOD-SCID mice receiving both 5×10⁶ and 5×10⁷ modK562 ECBNK cells when compared to those with no treatment (p=0.05, p=0.0007, respectively). There was no difference in survival when comparing mice that received 5×10⁶ vs 5×10⁷ modK562 ECBNK cells (p=0.0894) at 5 weeks. Tumor volume of mice receiving either dose of modK562 ECBNK cells was significantly less than those receiving WTK562 ECBNK cells (1.92±0.57 and 0.37±0.05 vs 3.41±0.25, p=0.0096 and p=0.0001, respectively).

CBMNCs stimulated and expanded with modK562 cells results in significant expansion of CBNK cells with enhanced in-vitro cytotoxicity, significant receptor expression of NK activation marker (LAMP-1), and perforin and granzyme B. Furthermore, modK562 ECBNK cells leads to increased survival and lower tumor burden of NOD-SCID mice xenografted with BL. Future directions include modK562 ECBNK cells to be genetically modified to express chimeric antigen receptor CD20 (MSCV-antiCD20-41BB-CD3 ζ) against CD20⁺ hematologic malignancies for future studies to evaluate whether targeting enhances in-vitro and in-vivo cytotoxicity.

No relevant conflicts of interest to declare.