Abstract
Natural killer (NK) cells are the major component of innate immunity with both cytotoxicity and cytokine producing effector functions. NK cells also regulate the interplay between innate immunity and adaptive immunity by secreting certain cytokines. Extrinsic regulators of NK cell development and function, including diverse ligands of NK cell receptors and cytokines from the microenvironment, have been extensively studied. However, intrinsic regulators for NK cell biology are still less understood. In our previous study on aggressive NK cell leukemia (ANKL), genomics and transcriptomics analyses indicated that c-MYC was universally upregulated and responsible for the proliferation and survival in ANKL cells (Manuscript in revision). Furthermore, STAT5, as a transcriptional regulator of c-MYC, was found to be essential in the survival and development of NK cells (Eckelhart et al., Blood 2011). In this regard, we want to understand the physiological and oncological roles of c-MYC in NK cells.
To achieve our goal, we made two mouse models including c-Myc loss-of-function (LOF) and c-Myc gain-of-function (GOF) in NK cells. Ncr1Cre knock-in mice, in which Cre recombinase was inserted into the Nkp46 locus, was used. We crossed the c-Mycf/f mice with Ncr1Cre mice to generate the NK cell specific c-Myc LOF model c-MycΔ/Δ/Ncr1Cre. To generate the c-Myc GOF model, we crossed the Tg(tetO-MYC) mice with Ncr1Cre and Rosa26-Loxp-Stop-Loxp(LSL)-rtTA-GFP mice to get the Tg(tetO-MYC)/ Ncr1Cre/LSL-rtTA-GFP (iMYC) mice, in which c-Myc expression can be induced in a doxycycline dependent manner in NK cells. c-MycΔ/Δ/Ncr1Cre mice were analyzed between 6 to 14-weeks old. iMYC mice were induced by doxycycline from 6-weeks old for over 2 months and then analyzed. Wild type littermates were used as controls. In both models, mice were born normally and showed no obvious difference in growth compared to their littermates.
In c-MycΔ/Δ/Ncr1Cre mice, a significant reduction of NK1.1+/DX5+ NK cell percentages in peripheral blood (3.6 ± 0.4% vs. 0.5 ± 0.1%, P < 0.0001, N=7) and spleen (2.4 ± 0.5% vs. 0.7 ± 0.1%, P < 0.01, N=6) was detected. In addition, the percentage of CD11b+ mature NK cells in the NK1.1+/DX5+ population was also reduced. In bone marrow (BM), although the total percentage of NK1.1+/DX5+ NK cells did not change, an obvious block of NK cell development was seen, as the majority of NK1.1+/DX5+ cells in BM were CD27+/CD11b- cells, which represent an immature pattern. To assess whether the NK cell proliferation is altered in this model, we performed BrdU labeling assays and found that BrdU incorporation rates decreased dramatically both in peripheral NK cells and BM NK progenitors. Functionally, we measured the IFN-γ secretion of splenic NK cells after PMA/Ionomycin stimulation. We found that the percentage of IFN-γ positive NK cells decreased significantly in c-MycΔ/Δ/Ncr1Cre mice (77.1 ± 7.0% vs. 53.8 ± 1.0%, N=3). Consistent with these data, the tumor surveillance was also severely impaired in this LOF model, as the number of lung metastatic sites significantly increased compared to the control mice in a B16F10 transplantation assay.
In contrast to the LOF model, in our GOF model, the NK1.1+/DX5+ NK cell number in peripheral blood increased (3.1 ± 0.2% vs. 4.4 ± 0.5%, P < 0.05, N=7). Additionally, a small increase in the percentage of CD27-/CD11b+ population in NK1.1+/DX5+ cells was seen. Interestingly, however, the ability of IFN-γ secretion of splenic NK cells after PMA/Ionomycin stimulation was decreased in iMYC mice (72.8 ± 0.3% vs. 58.7 ± 1.1%, N=3), which showed the same alteration observed in c-MycΔ/Δ/Ncr1Cre mice. This result is probably not due to the impaired maturation, but rather it is the result of the higher percentage of CD27-/CD11b+ cells, which were considered terminally differentiated NK cells with lower cytotoxic functions.
In summary, we found that c-Myc is essential for NK cell development, proliferation, and tumor surveillance. NK cell maturation and proliferation were impaired in the c-Myc LOF models and were boosted in the c-Myc GOF models. Our results also provide a mechanism basis for the potential application of targeting c-Myc in NK cells ex vivo or in vivo expansion, and NK-mediated immunotherapy. Future studies are needed to delineate the underlying mechanisms and explore the applications.
Disclosures
No relevant conflicts of interest to declare.
Differential controls of MAIT cell effector polarization by mTORC1/mTORC2 via integrating cytokine and costimulatory signals
