5, 13, 14)

5, 13, 14). Tfpi of MULT1, an NKG2DL, that correlates with an induction of DNA damage. Treatment with the ATM DNA damage restoration pathway inhibitor KU55933 (KU) during activation reduced NCE by improving manifestation of activation markers and genes involved in cell survival, through sustaining NKG2D manifestation and conserving cell features. Importantly, NK cells expanded ex lover vivo in the presence of KU displayed improved antitumor effectiveness in both NKG2D-dependent and -self-employed mouse models. Collectively, these data demonstrate that NCE is definitely caused by DNA damage and is controlled, at least in part, by NKG2D. Further, the prevention of NCE is definitely a promising strategy to improve NK cellCbased immunotherapy. < 0.05, **< 0.01, ***< 0.001). No significant variations were found when comparisons among the control, resolved, and chronic organizations were made. We next evaluated the effect of in vivo chronic activation with IL-2 or Poly I:C, directly or indirectly activating NK cells, respectively, on NCE to be able to demonstrate that any type of chronic stimulation results in the induction of NCE. Furthermore, Poly I:C, a known ligand for TLR-3, was used like a model to mimic viral illness (21). Both models showed similar results obtained with the IL-15 model (Number 1), concerning the NK cell phenotype and features (Supplemental Number 2, ACC; supplemental material available on-line with this short article; https://doi.org/10.1172/jci.insight.127729DS1). A reduction of inflammatory cytokines in the serum of chronically treated organizations was also observed in some, if not all, of the models evaluated; among these, IFN- was a cytokine that was downregulated in all of them (Supplemental Number 2D). Noticeably, chronic activation also caused an increase of the immature-like NK cell subset (CD27+CD11bC) in the BM that could justify the consistent reduction of total numbers of splenic NK cells observed in all the models (Supplemental Number 3, A and B) and suggest a reduced ability of NK cells to respond to stimuli and/or improved cell death. The lack of response to cytokine activation after chronic exposure in the models evaluated was not explained Asimadoline by changes in the expression of IL-2 receptor (IL2R), an important component in the signaling through IL-2 and IL-15, as others have Asimadoline also reported (ref. 18 and Supplemental Physique 3C). In order to identify an exhaustion phenotypic signature, we then analyzed the data originating from the 3 different in vivo models using principal component analysis (PCA; Physique 1, E and F). PC1 was able to efficiently individual acutely stimulated murine NK cells from control mice (control and resolved groups) and was highly influenced by levels of expression of activation markers, such as Thy1.2, and inhibitory and activating receptors, such as DNAM1 or TIGIT (Physique 1, Asimadoline E and G). Conversely, PC2 provided a better separation between the chronically and acutely stimulated groups. As shown in Physique 1, E and G, Eomes and NKG2D downregulation and KLRG1 and NKG2A upregulation upon chronic activation were major factors segregating mice and conditions along the PC2 axis. The variability in the expression of other activating and inhibitory receptors between the models, such as with DNAM1 (Physique 1A and Supplemental Physique 2A), TIGIT (Supplemental Physique 4, A and B), Tim3 (Supplemental Physique 4, C and D), or PD1 (Supplemental Physique 4, E and F), did not account for the results obtained in the PCA, whereas the changing patterns of NKG2D (Physique 1A; Supplemental Physique 2A; and Supplemental Physique 4, G and H), Eomes, KLRG1, and NKG2A (Physique 1A and Supplemental Physique 2A) were comparable in all the models. Asimadoline These data also suggest that changes of the expression of these markers are more suitable in identifying exhaustion across models. To validate this phenotypic signature, we used a more physiological model.