Natural killer (NK) cells are a subset of lymphocytes that play a critical role in both innate and adaptive immune responses and provide defense against microbial infections and malignant transformation. When stimulated through their activating or cytokine receptors, NK cells rapidly produce cytokines and chemokines, including IFNγ, TNFα, TGFβ, GM-CSF, MIP1α, MIP1β, IL-10, and others, which can affect the function of hematopoietic stem cells (HSC)s. We hypothesized that NK-mediated cytokine production could regulate HSC function and that its deregulation may favor HSC malignant transformation and proliferation. In this regard, a few studies have shown that NK cells are defective in stem cell-derived diseases such as chronic myeloid leukemia (CML), for example, raising the hypothesis that immunological dysfunctions may be involved in initiating or contributing to the pathogenesis of myeloproliferative diseases. In contrast to CML, the role of NK cells in BCR-ABL1 negative-myeloproliferative neoplasms (MPN) is currently unexplored. Therefore, we aimed to study NK cell activity in murine and human primary cells from MPN. NK cells and receptors were quantified by flow cytometry by the use of spleen cells from a conditional knockin Jak2V617F (Jak2VF) murine model, that faithfully resemble the main clinical and laboratorial characteristics of human Polycytemia Vera (PV), and from MPN patients diagnosed and followed at the Clinical Hospital of the Medical School of Ribeirao Preto, University of Sao Paulo. For immunophenotyping of murine lymphocytes, spleen cells were isolated and stained with fluorescence-conjugated antibodies against murine Ter119, CD19, CD4, CD8, CD3, NK1.1 (for NK cells quantification) and NKG2A, NKG2D, Ly49H and Ly49D (for receptors analysis). For analysis of human lymphocytes, peripheral blood mononuclear cells were purified by density gradient centrifugation, resuspended in serum-enriched phosphate buffered saline and stained with fluorescence-labeled anti-human antibodies against the following molecules: CD45, CD19, CD3, CD16, CD56. All the stainings were performed for 20 minutes at 40 C and cells were acquired using a FACScalibur flow cytometer (BD Biosciences). Data was analyzed by the use of the Flojo software version X and the statistical analyses were performed by the use of SPSS version 20. When compared to Jak2WT controls, Jak2VF mice had lower numbers of NK cells (2.09% versus 22.7% of the spleen cells, P = 0.022) and NKG2A (0.57% versus 2.47% of NK cells, P = 0.037) and NKG2D (0.17% versus 0.46% of NK cells, P = 0.020) receptors. For the study of human NK cells, 38 patients diagnosed with MPN and who were not in use of cytoreductive therapy were included and classified according to the WHO 2008 criteria as Polycytemia Vera (PV, n = 11), Essential Thrombocytemia (ET, n = 11), Primary Myelofibrosis (PMF, n = 11) or Unclassified MPN (uMPN, n = 5). Blood samples from 34 healthy donors served as controls. JAK2V617F (JAK2VF) mutated patients had lower numbers of CD16+ CD56dim NK cells as compared to JAK2VF negative and healthy controls (0.73% versus 1.06% versus 1.54%, P = 0.038, respectively). Among the JAK2 mutated patients, those with PV and PMF presented lower numbers of NK cells than ET patients. In agreement, NK cells were reduced in JAK2VF PV patients when compared to JAK2VF negative and controls (0.39% versus 1.42% and 1.54%, P = 0.047). PMF patients had lower NK cell numbers than controls, regardless of JAK2VF mutation (0.35%, 0.49% and 1.54% for JAK2VF positive, negative and controls, respectively, P = 0.025). No significant differences were observed in ET or uMPN. In summary, NK cells were reduced among lymphocytes in MPN patients, especially in the JAK2VF mutated cases, as shown in PV patients and Jak2VF mice. In addition, the expression of NK receptors was reduced in Jak2VF mice, thus suggesting that these cells are defective in MPN. New insights into the pathogenesis of this group of diseases are of great relevance considering that MPN treatment remains non-curative with the strategies available so far despite the development of drugs that inhibit JAK2. Thus, identifying the mechanisms of NK-cell deficiency in MPN and the detailed interaction between these cells and the malignant stem cell may help to establish combination therapeutic strategies that may improve therapy responses in MPN. Disclosures No relevant conflicts of interest to declare.

In cancer, tumor cells and their neoplastic microenvironment can sculpt the immunogenic phenotype of a developing tumor. In this context, natural killer (NK) cells are subtypes of lymphocytes of the innate immune system recognized for their potential to eliminate neoplastic cells, not only through direct cytolytic activity but also by favoring the development of an adaptive antitumor immune response. Even though the protective effect against leukemia due to NK-cell alloreactivity mediated by the absence of the KIR-ligand has already been shown, and some data on the role of NK cells in myeloproliferative neoplasms (MPN) has been explored, their mechanisms of immune escape have not been fully investigated. It is still unclear whether NK cells can affect the biology of BCR-ABL1-negative MPN and which mechanisms are involved in the control of leukemic stem cell expansion. Aiming to investigate the potential contribution of NK cells to the pathogenesis of MPN, we characterized the frequency, receptor expression, maturation profile, and function of NK cells from a conditional Jak2V617F murine transgenic model, which faithfully resembles the main clinical and laboratory characteristics of human polycythemia vera, and MPN patients. Immunophenotypic analysis was performed to characterize NK frequency, their subtypes, and receptor expression in both mutated and wild-type samples. We observed a higher frequency of total NK cells in JAK2V617F mutated MPN and a maturation arrest that resulted in low-numbered mature CD11b+ NK cells and increased immature secretory CD27+ cells in both human and murine mutated samples. In agreement, inhibitory receptors were more expressed in MPN. NK cells from Jak2V617F mice presented a lower potential for proliferation and activation than wild-type NK cells. Colonies generated by murine hematopoietic stem cells (HSC) after mutated or wild-type NK co-culture exposure demonstrated that NK cells from Jak2V617F mice were deficient in regulating differentiation and clonogenic capacity. In conclusion, our findings suggest that NK cells have an immature profile with deficient cytotoxicity that may lead to impaired tumor surveillance in MPN. These data provide a new perspective on the behavior of NK cells in the context of myeloid malignancies and can contribute to the development of new therapeutic strategies, targeting onco-inflammatory pathways that can potentially control transformed HSCs.

C/EBPs (CCAAT/enhance-binding proteins) are a family of transcription factors involved in a variety of hematopoietic processes, regulating both terminal differentiation and cellular proliferation. Among these, it was previously reported that C/EBP gamma (C/EBPg) has a role in the development of Natural Killer (NK) cells. However, the mechanisms of such regulation are unknown. NK cells are lymphocytes with effector functions of cytotoxicity and production of cytokines, both dependent on a dynamic equilibrium between the expression of activating and inhibitory receptors as well as cytokine receptors. The two functions (cytotoxic and secretory) make NK cells important components of hematopoiesis, able to eliminate susceptible targets as well as recruit other cells to amplify inflammatory responses. With the aim of studying the regulation of NK cells by C/EBPg, we isolated NK cells from transgenic Cebpg knockout (KO) mice and controls to analyze their function. To characterize NK cells, we analyzed their frequency (Lineage-/CD3-/NK1.1+ cells) and the expression of the receptors NKG2D, Ly49D and NKG2A by flow cytometry of splenocytes. Both analyses showed no difference between control or Cebpg KO NK cells. Although the numbers of NK cells and their receptors were similar between Cebpg WT and KO animals, a functional assay that measured NK cell degranulation by CD107a expression after co-incubation with YAC-1 target cells showed that the expression of this marker was 5-times lower in Cebpg KO splenocytes than in controls (CT = 12.44 ± 2.50%; KO = 2.255 ± 0.67%, p=0.007), suggesting that Cebpg deficient NK cells are not fully activated after target cell recognition. In addition, a cytotoxicity assay by flow cytometry was performed using a fluorescent probe (Cell Tracker Orange) that was incorporated to YAC-1 cells upon exposure to sorted and IL-2 activated NK cells in culture. In the 10:1 NK:target cells ratio, Cebpg KO cells were significantly less cytotoxic than NK control cells (CT = 23.36 ± 8.67%; KO = 10.60 ± 1.66%, p=0.038). The other NK:target cells ratios of 5:1 and 1:1 showed the same tendency. In addition, the functional subtypes of these cells were characterized according to the expression of CD27 and CD11b, which allowed the identification of NK subpopulations as immature secretory, mature secretory, cytotoxic or tolerant. The KO animals showed higher percentages of secretory cells (CT = 10.77 ± 5.38%; KO = 12.98 ± 13.63%, p=0.0002) and a reduction of cytotoxic cells in comparison to the NK control cells (CT = 12.22 ± 11.08%; KO = 10.65 ± 3.82% p=0.013). Cytokine levels of IL-2, IL-4, IL-6, IL-10, IL-17α, TNFα and IFNγ, obtained from NK culture supernatants, were measured by flow cytometry, after IL-2 activation. Among these cytokines, the production of IFNγ by Cebpg-deficient NK cells was reduced (CT = 37.68 ± 0.51 pg/mL; KO = 22.34 ± 0.14 pg/mL, p=0.023). Together, these experiments indicate that C/EBPg regulates NK cell cytotoxicity. This may be explained, at least in part, by the reduced frequency of the mature cytotoxic NK subpopulation as compared to the secretory subtypes. Moreover, IFNγ may be an important target for the regulation of NK cell function. Finally, C/EBPg seems to be critical to mediate NK cell functions and not only for their development from the ontogenetic point of view. Disclosures No relevant conflicts of interest to declare.