Search

Innate immune cells [Natural killer (NK) and gamma-delta (γδ) T-cells] have the advantage of mediating graft versus leukemia (GVL) without graft versus host disease (GVHD). Therefore, the infusion of activated innate immune cells post allogenic hematopoietic stem transplant (AHSCT) is a promising adoptive immunotherapy strategy for relapsed and/or refractory myeloid malignancies. Microbead depletion of T-cells and B-cells has been used as a graft manipulation method to prevent GVHD post haploidentical AHSCT. These grafts are enriched for NK and γδ T-cell receptor (TCR

The derivation of natural killer (NK) cells from human pluripotent stem cells (hPSCs) is an exciting alternative for adoptive immunotherapy strategies in the treatment of cancer. While NK cells are component of the embryonic hematopoietic repertoire, their ontogenic origins remain unclear. During mammalian embryogenesis, hematopoietic development in the early yolk sac consists of the (HSC)-independent primitive and erythro-myeloid progenitor (EMP) hematopoietic programs. Neither population is known to harbor lymphoid potential. As we previously observed NK cell development from extra-embryonic-like progenitors derived from hPSCs, we asked whether a yolk sac-derived program possesses NK cell potential. Surprisingly, we determined that EMPs isolated from the yolk sacs of E9.5 mouse embryos harbor robust NK cell potential. Further, these NK cells exhibited a potent degranulation response following stimulation, making them functionally distinct from HSC-derived NK cells. In parallel studies of hPSCs differentiated as previously described, we found that extra-embryonic-like CD34+ myeloid progenitors, similar to murine EMPs, give rise to NK cells that are biased for cytolytic degranulation. In contrast, hPSC-derived CD34+ lymphoid progenitors give rise to NK cells that are phenotypically similar to those derived from cord blood, since both exhibit a poor degranulation response but robustly produce inflammatory cytokines. Collectively, our studies in mouse embryos and human PSCs point to a novel, hematopoietic stem cell (HSC)-independent origin of embryonic NK cells that differ functionally from HSC-derived NK cells. These findings not only identify a myeloid origin to embryonic NK cells, but also have important implications for the design of hPSC-derived NK cell-based therapeutics.

NK cells develop in the bone marrow and complete maturation in peripheral organs, but the factors controlling maturation are incompletely understood. The miR-15/16 family is highly expressed in NK cells, and to study its role in NK cells, we generated a novel mouse model of NK-specific miR-15/16 deficiency (15aKO). 15aKO NK cells have a deficiency of mature (CD11b+CD27-) NK cells, with no alterations in total NK cell numbers or death, consistent with a block in maturation. The transcription factor Myb is expressed highly in immature NK cells, nearly absent in mature NK cells, and has been implicated in NK maturation. We found that miR-15/16 represses Myb, and Myb levels are increased in 15aKO NK cells. To directly investigate the role of Myb in NK cells, we developed a novel NK maturation system in which immature primary NK cells are transduced, adoptively transferred, and matured in vivo. As in the 15aKO mouse, transferred 15aKO NK cells had defective maturation. However, both 1) restoration of miR-15/16, or 2) knockdown of Myb, rescued maturation of the 15aKO NK cells, demonstrating that miR-15/16’s repression of Myb is critical to proper NK cell maturation. Furthermore, WT NK cells overexpressing Myb did not mature as well as control (WT NK+GFP) cells. Thus, the regulation of Myb by miR-15/16 is essential for NK cell maturation, revealing a novel mechanism controlling NK cell development.

Phosphatase and tensin homolog (PTEN) is a critical negative regulator of the PI3 kinase pathway, members of which are essential for regulating natural killer (NK) cell signaling and activation. To evaluate whether PTEN was required for normal NK cell functions, we utilized a mouse model of NK specific PTEN deficiency (Ncr1iCre knockin x PTENflox). PTEN deletion resulted in significant loss of NK cells in the bone marrow and other lymphoid tissues, but markedly increased numbers within the peripheral blood and lung. Surprisingly, we observed near equivalent NK cell maturation within peripheral organs regardless of PTEN expression, suggesting that PTEN operates to re-distribute NK cells during development. Similarly, despite impaired migration to chemoattractants in vitro, PTEN-deficient NK cells egress more efficiently from the bone marrow and preferentially reside in sinusoidal compartments, but are specifically retained in the lung and peripheral vasculature. Given the inappropriate localization in the absence of PTEN, we further evaluated whether cytotoxic functions were altered. While PTEN deficient NK cells have enhanced cytotoxicity against co-localized RMAS tumors, they are unable to migrate to distal tumor sites. These data suggest that PTEN is required for the trafficking of NK cells during both homeostasis and in the presence of malignancy, during which it represses cytotoxic functions.

The heterotrimeric high affinity IL-2 receptor (IL-2Rαβγ) is expressed on CD56bright human NK cells and confers responsiveness to picomolar concentrations of IL-2. In contrast, resting CD56dim NK cells lack IL-2Rα (CD25), but constitutively express the intermediate affinity IL-2Rβγ chains. Here, we show that cytokine-activation induces rapid and prolonged expression of CD25 on CD56dim NK cells, especially the combination of IL-15 plus IL-18. This expression of CD25 correlated with STAT5 phosphorylation in response to picomolar concentrations of IL-2, indicating the presence of a signal-competent IL-2Rαβγ. Furthermore, picomolar concentrations of IL-2 acted synergistically with IL-12 to co-stimulate IFN-γ production by pre-activated CD56dim NK cells. Similarly, low dose IL-2 enhanced the cytotoxicity of pre-activated CD56dim NK cells against K562 tumor targets. Both effects were dependent on the expression of CD25. Preliminary experiments also suggest that the induction of CD25 on CD56dim NK cells enhances proliferative capacity and survival in cultures supplemented with low dose IL-2. Since CD56dim NK cells represent the major subset of human peripheral blood NK cells, these data support a rationale for immunotherapeutic strategies that include cytokine pre-activation prior to adoptive NK cell transfer, followed by low dose IL-2 therapy.

MiR-155 is a well-documented small RNA regulator of a variety of immune cells, including adaptive T and B lymphocytes. However, its role in regulating the innate NK cell response is unclear. We evaluated the expression of miR-155 using real-time qPCR and found that miR-155 is expressed at modest levels in resting mouse and human NK cells, but is upregulated upon stimulation, especially with IL-12+15+18 (30-fold). Therefore, we hypothesized that miR-155 regulates NK cell function following activation. We utilized miR-155-/- mice (155KO), NK-specific miR-155 overexpression mice (155OE), and miR-155 lentiviral transduction to examine the role of miR-155 in NK cell functional responses. miR-155KO, miR-155OE, and miR-155-transduced NK cells were all found to secrete more IFN-γ (by ELISA) upon ligation-mediated or cytokine stimulation, with no apparent defects in maturation or development. When further dissected by flow cytometry, we found that 155OE and transduced NK cells had increased IFN-γ MFI, indicating more “per-cell” production. In contrast, the 155KO NK cells had a higher percentage of IFN-γ+ NK cells, with similar MFI, indicating a potential activation threshold defect. Though both 155KO and 155OE NK cells produce more IFN-γ, the cellular basis for each, i.e. miR-155 target, likely differs. Therefore, we used unbiased RISC-IP to identify relevant miR-155 targets, which included previously validated miR-155 targets (e.g., SHIP-1) as well as novel miR-155 NK cell targets.