bims-nakice Biomed News
on Natural killer cells
Issue of 2026–09–27
five papers selected by
Santosh Phuyal, Oslo Universitetssykehus



  1. Cell Death Dis. 2026 Jul 24. pii: 820. [Epub ahead of print]17(1):
      Natural killer (NK) cells are critical for both innate and adaptive immunity, and their effective regulation holds significant promise for tumor immunotherapy. The terminal maturation of NK cells is closely associated with enhanced effector functions, but the role of epigenetic mechanisms in this process remains incompletely understood. Here, we identify the histone methyltransferase EHMT2 as an epigenetic regulator that restrains NK cell maturation and function. EHMT2 expression is markedly downregulated upon cytokine activation in human NK cells and inversely correlated with effector gene expression across tissues and cancer types. NK cell-specific deletion of Ehmt2 in mice promotes terminal NK cell maturation, augmenting IFN-γ production and cytotoxicity both in vitro and in vivo. Integrated RNA-seq and H3K9me2 CUT&Tag profiling reveal that Ehmt2 loss reduces the repressive histone mark H3K9me2 and derepresses transcriptional programs associated with NK cell differentiation and cytotoxicity. Pharmacological inhibition of EHMT2 recapitulated these effects in both human umbilical cord blood (hUCB)-hematopoietic stem cell (HSC)-derived NK cells and peripheral blood NK cells, promoting terminal maturation and tumor cell killing. These findings define EHMT2-dependent H3K9me2 deposition as a key epigenetic barrier to NK cell effector programming and suggest that targeting EHMT2 may potentiate NK cell-based immunotherapies.
    DOI:  https://doi.org/10.1038/s41419-026-09124-y
  2. Immunology. 2026 Sep 20.
      Natural killer (NK) cells are innate immune cells that eliminate virus-infected and tumour cells by a plethora of signalling pathways, involving the production of cytokines and cytotoxic molecules. NK cell activation and functions demand substantial energy and fatty acids are energy-rich molecules reported to be important for NK cell functions upon viral infections. However, it remains unclear whether NK cells preferentially take up a specific fatty acid. Here, we show in the acute Friend retrovirus infection of mice that linoleic acid (LA) exerts a dual role in modulating NK cells. Interestingly, NK cells increased the uptake of LA upon virus infection in vitro and in vivo, with higher LA uptake correlating with increased activation of NK cells. While ex vivo stimulation with LA increased the expression of cytolytic molecules and subsequently elicited increased target cell killing, it paradoxically decreased the NK cell proliferative capacity, mitochondrial metabolism and increased the cellular stress. To further elucidate the role of LA in vivo, we treated mice with exogenous LA during retrovirus infection. While we observed a significant increase of LA levels in the spleen, splenic NK cell activation, cytotoxicity and viral loads were not altered by LA therapy. In addition, we found a decrease in mitochondrial mass and energy levels in NK cells after LA treatment, reflecting its strong effect on mitochondrial metabolism. Our results demonstrate the importance and direct impact of LA for NK cell functions and their metabolic profile. The dual role of LA in enhancing NK cell responses while compromising mitochondrial integrity highlights its potential as a therapeutic adjuvant, alongside other fatty acids, to modulate NK cell functions.
    Keywords:  fatty acids; linoleic acid; metabolism; natural killer cells; retrovirus infection
    DOI:  https://doi.org/10.1111/imm.70205
  3. Nature. 2026 Sep;657(8133): 870-872
      
    Keywords:  Cancer; Cell biology; Molecular biology
    DOI:  https://doi.org/10.1038/d41586-026-02929-z
  4. Proc Natl Acad Sci U S A. 2026 Sep 29. 123(39): e2612523123
      Pancreatic ductal adenocarcinoma (PDAC) is among the most hypoxic human tumors. Because fatty acid (FA) desaturation is oxygen-dependent, hypoxia can limit monounsaturated FA (MUFA) production, increase membrane lipid saturation, and activate endoplasmic reticulum stress responses, including IRE1α-XBP1s. Here, we found that under oxygen- and MUFA-limiting conditions, spliced XBP1 (XBP1s) is upregulated but unexpectedly exerts a cytotoxic rather than cytoprotective role in PDAC cells. This effect did not differ substantially between classical and basal subtypes. In contrast, pharmacologic or genetic XBP1s inhibition had limited effects on tumor growth and apoptosis in vivo, suggesting this cytotoxicity is largely bypassed by factors in the tumor microenvironment. Consistent with our previous findings that cancer-associated fibroblasts supply unsaturated lipids to tumor cells, subcutaneous tumors showed abundant alpha-smooth muscle actin (α-SMA)-positive stroma, supporting the possibility that stromal lipid supply protects tumors from XBP1s-dependent lipotoxicity. Although XBP1s expression increased during PDAC progression, its distribution remained focal and heterogeneous within human tumors, suggesting spatially restricted IRE1α-XBP1s pathway activation that may limit the efficacy of monotherapy in patients. However, MRTX1133-resistant PDAC became more susceptible to IRE1α-XBP1s targeting, and MRTX1133 acutely activated this pathway in parental cells upon treatment. Importantly, the IRE1α RNase inhibitor B-I09 clearly synergized with MRTX1133 in vitro and in vivo, moreover, this is likely due to MYC-fatty acid synthase (FASN) dysregulation. Together, these findings identify context-dependent vulnerabilities of the IRE1α-XBP1s pathway in PDAC and provide a rationale for combining inhibition of IRE1α and KRAS to enhance therapeutic responses.
    Keywords:  ER stress responses; KRAS inhibitors; hypoxia; lipid metabolism
    DOI:  https://doi.org/10.1073/pnas.2612523123
  5. Commun Biol. 2026 09 19. pii: 1221. [Epub ahead of print]9(1):
      Mammalian gene switches enable programmable cell behavior. However, current switches on transcriptional and translational layers require de novo RNA and protein synthesis, and secreted outputs must additionally undergo folding, post-translational processing, and intracellular trafficking, imposing delays that limit rapid extracellular responses. Rapid secretion-control switches instead act on pre-synthesized proteins by controlling retention, trafficking, storage, or release. Most current platforms exploit the classical endoplasmic reticulum (ER)-Golgi pathway, including engineered stimulus-secretion coupling in specialized secretory cells and ER retention, retrieval-signal cleavage, or synchronized trafficking in general mammalian hosts. Although these strategies improve response kinetics, they remain limited by ER dependence, host-cell specificity, cargo compatibility, basal leakage, and post-release transport delays. Here, we review current secretion-control architectures and highlight unconventional protein secretion as an underexplored source of design principles for positioning regulatory control closer to terminal protein export and expanding the architectures available for mammalian secretion control.
    DOI:  https://doi.org/10.1038/s42003-026-10984-5