Cell Cycle Timing Shapes How Natural Killer Cells Fire Their Weapons

Natural killer cells are among the most rapid responders of the human immune system, patrolling the bloodstream and tissues for cells that show signs of viral infection or malignant transformation. Unlike T cells, which require days of clonal expansion before they become fully armed, NK cells can deploy cytotoxic machinery and inflammatory signals within minutes of encountering a target. Yet even the fastest weapons in biology obey internal clocks. A new computational and experimental study published in PLOS Computational Biology reports that the signaling responses triggered by one of the NK cell’s key activating receptors vary systematically depending on where the cell sits in its division cycle, a finding that reshapes how immunologists think about receptor signaling in these innate lymphocytes.

The research team, led by Darren Wethington and Jayajit Das and including investigators from the laboratory of Lewis Lanier as well as collaborators at Stanford and Ohio State, focused on NKG2D, an activating receptor that recognizes stress-induced ligands frequently displayed by tumor cells and infected cells. When NKG2D engages its ligands, a cascade of phosphorylation events sweeps through the cell within minutes, activating signaling proteins such as Vav1, Erk, and the ribosomal protein S6. These rapid biochemical changes ultimately determine whether the cell releases its cytotoxic granules, produces cytokines, or proceeds to divide. Previous work using single-cell mass cytometry in macrophages had hinted that the abundance of phosphorylated signaling proteins fluctuates across the cell cycle, suggesting that signaling kinetics might be intertwined with the much slower process of cell cycle progression, which unfolds over hours rather than minutes.

To test this idea in NK cells, the researchers performed longitudinal CyTOF measurements on primary human CD56-dim NK cells that had been treated with the cytokine IL-2 and then stimulated through NKG2D. CyTOF, or mass cytometry, allows dozens of proteins to be quantified in individual cells simultaneously by tagging antibodies with heavy metal isotopes and reading them off with a mass spectrometer. By measuring signaling protein phosphorylation at multiple time points stretching up to 256 minutes after stimulation, the team captured the full arc of the signaling response with single-cell resolution, while simultaneously tracking markers that identify each cell’s position in the cell cycle.

The resulting data revealed striking stage-specific differences. Phosphorylated Vav1, phosphorylated Erk, and phosphorylated S6 all showed kinetics that depended on the cell’s cycle stage, with distinct amplitude and timing profiles across G0, G1, S, and the combined G2 plus M phases. Most notably, CD107a, a surface marker of degranulation that serves as a widely used proxy for the cell’s cytotoxic output, showed its largest upregulation specifically in cells in the G2 plus M stage. In other words, NK cells poised near mitosis were the ones most prepared to unleash their killing machinery upon NKG2D engagement, while cells in other stages mounted measurably weaker or differently timed responses.

This observation raises a fundamental question about cause and effect. The cell cycle is an order of magnitude slower than receptor signaling, operating on a timescale of several hours, while phosphorylation events resolve within minutes to a few hours. One might expect the fast process to simply ride on top of the slow one, with protein abundance differences across cycle stages passively reflecting changes in the expression of signaling components. The researchers instead built stochastic minimal computational models that couple the signaling network to the cell cycle machinery, allowing them to ask whether active modulation by cell cycle progression was required to reproduce the observed data.

The modeling results pointed firmly toward active modulation. Stochastic models that coupled signaling and cell cycle processes could account for the stage-specific kinetics of pVav1, pErk, and pS6, whereas models lacking this coupling could not. In practical terms, the analysis suggests that the cell cycle does not merely passively supply different starting amounts of signaling proteins; rather, the progression through cycle stages actively reshapes the speed and magnitude of the receptor-induced phosphorylation cascades. Because the models are stochastic, they also capture the cell-to-cell variability that is inherent in single-cell measurements, providing a quantitative bridge between population-level averages and the heterogeneous behavior of individual lymphocytes.

The implications extend beyond basic cell biology. NK cell-based cancer immunotherapies, including adoptive transfer of ex vivo expanded NK cells and engineered NK lines, depend on maximizing the cytotoxic potency of the transferred cells. If degranulation capacity peaks in the G2 plus M stage, then the proliferative state of an NK cell product at the moment of infusion or target encounter could materially influence its antitumor activity. The study’s framework offers a way to think quantitatively about this dependence, and its authors note that the modeling approach can be extended to investigate the interplay between receptor signaling and cell cycle progression in other cell types, where similar coupling may govern effector functions ranging from macrophage inflammatory responses to T cell activation.

Methodologically, the work demonstrates the power of combining longitudinal single-cell proteomics with mechanistic modeling. Static snapshot measurements of signaling would have averaged over cells in different cycle stages and obscured the stage-specific patterns entirely. By following stimulated populations over nearly four and a half hours and annotating each cell with cell cycle markers, the researchers could disentangle a slow variable, cycle stage, from fast signaling dynamics. The minimal stochastic modeling strategy keeps the number of fitted parameters manageable while still resolving which interactions are necessary to explain the data, an approach that balances mechanistic insight with statistical tractability.

For the broader field of immunology, the study adds NK cells to a growing list of immune cell types whose signaling behavior cannot be fully understood without reference to the cell cycle. It also underscores a conceptual shift in systems immunology: signaling networks and the cell cycle are not independent modules but coupled dynamical systems, each modulating the other. As single-cell technologies continue to mature, the expectation is that such coupled models will become standard tools for interpreting immune cell behavior, potentially guiding the design of cell therapies in which the timing of stimulation relative to the division cycle becomes a deliberate engineering variable rather than an uncontrolled source of variability.

The research, published on October 13, 2026, in PLOS Computational Biology, provides both a rich experimental dataset and a reusable computational framework. Its central message is deceptively simple: when an NK cell decides to kill, the decision is shaped not only by the receptors on its surface and the ligands on its target, but by the ancient internal program that governs when the cell will next divide. Understanding that three-way conversation among receptor, target, and cell cycle may prove essential for predicting when NK cells strike effectively against cancer and infection, and for designing the next generation of NK cell immunotherapies.

Subject of Research: Cell cycle dependence of natural killer cell receptor signaling and activation

Article Title: Natural killer cell receptor signaling and activation depend on cell cycle stages

Article References: Wethington, D., Nayak, I., Jensen, H., Stewart, W. C., Aguilar, O. A., Chen, S.-Y., Nolan, G. P., Behbehani, G. K., Lanier, L. L., & Das, J. (2026). Natural killer cell receptor signaling and activation depend on cell cycle stages. PLOS Computational Biology, 22(10), e1014825. https://doi.org/10.1371/journal.pcbi.1014825

Image Credits: AI Generated

DOI: 10.1371/journal.pcbi.1014825

Keywords: natural killer cells, NKG2D, cell cycle, CyTOF, mass cytometry, signaling kinetics, pVav1, pErk, pS6, CD107a, degranulation, computational modeling