Below the NK-cell surface: PIP2

In this issue Blood, Micucci and colleagues demonstrate that PIP2 plays an essential role in NK-cell cytotoxicity by controlling the release of lytic mediators at the NK cell–target interface.

Membrane phospholipids are essential for a variety of cellular functions. They act as second messengers that couple engagement of cell-surface receptors with recruitment and activation of downstream intracellular molecules, and are also involved in cell adhesion, motility, actin cytoskeleton dynamics, phagocytosis, vesicle trafficking, endocytosis, and exocytosis.^1-4  Despite their established physiological relevance, membrane phospholipids have generally been a rather obscure topic in the context of natural killer (NK)–cell function. Now, Micucci and colleagues bring phosphatidylinositol 4,5-bisphosphate (PIP2) to the forefront of NK-cell biology in this issue of Blood.

PIP2 is the precursor of 2 second messengers that are indispensable for cellular signaling—inositol 1,4,5-trisphosphate (IP3), which mobilizes Ca2⁺ from intracellular stores, and diacylglycerol (DAG), which activates protein kinase C (PKC). Cleavage of PIP2 into IP3 and DAG is mediated by members of the phospholipase C (PLC) family.⁵  PIP2 also interacts with the actin-binding proteins gelsolin and profilin, as well as with many other signaling molecules that contain pleckstrin homology (PH) domains.^1-4  To visualize PIP2, Micucci et al transfected a human NK-cell line with a lentivirus encoding a chimeric protein consisting of the pleckstrin homology (PH) domain of PLCδ1 fused to green fluorescent protein (GFP).⁶  Because the PH domain of PLCδ1 binds PIP2, the PLCδ1-GFP reporter allows confocal microscopy imaging of PIP2 at the site of its production in living cells. Moreover, the PLCδ1-GFP reporter can act in a dominant-negative manner by sequestering PIP2 from its physiological targets. Micucci et al found that in resting NK cells, the PLCδ1-GFP reporter was immediately detectable in small clusters at the cell membrane. In contrast, when NK cells interacted with a target cell, the PLCδ1-GFP reporter disappeared from the NK-cell–target-cell synapse, suggesting breakdown of PIP2 for the generation of second messengers. Furthermore, sequestration of PIP2 by the PLCδ1-GFP reporter prevented exocytosis of NK-cell lytic granules and, consequently, lysis of target cells.

To corroborate the assumption that PIP2 is required for NK cell–mediated cytotoxicity, Micucci and colleagues analyzed the effect of blocking PIP2 synthesis. PIP2 is generated from phosphatidylinositol-4-phosphate (PI4P) by PI4P-5 kinases.¹  Of the 3 PI4P-5 kinase isoforms that have been identified, NK cells preferentially express PI4P-5KIα and PI4P-5KIγ. Therefore, to effectively prevent generation of PIP2, Micucci et al inhibited expression of PI4P-5KIα or PI4-P5KIγ by short hairpin RNA (shRNA). Down-regulation of either enzyme impaired lysis of tumor targets by NK cells, whereas secretion of IFNγ was unaffected. NK cell–mediated cytotoxicity is a sequential process involving conjugation of NK cells with the target cell, polarization of lytic granules toward the target cell, and exocytosis of the lytic mediators. Micucci and colleagues found that PIP2 is selectively required for exocytosis of lytic granules. In a previous study, the same group demonstrated that NK cell–mediated cytotoxicity requires the small GTPase ARF6,⁷  which mediates membrane recruitment and activation of PI4–5KIs.⁸  Collectively, these studies conclusively demonstrate that PIP2 is re-quired for NK cell–mediated cytotoxicity, specifically for the exocytosis of lytic granules. This conclusion probably applies not only to NK cells, but also to all cytotoxic lymphocytes, including CD8 T cells.

Intriguingly, Micucci et al found that while inhibition of PI4P-5KI repressed the activity of PLCγ and consequent generation of IP3, it had no effect on the activity of PI3K. This is surprising, given that PI3K uses PIP2 as a substrate to generate PIP3, another key regulator of cytotoxicity and NK-cell function.⁹  The data reported by Micucci et al thereby suggest that PIP2 is generated in distinct cell compartments and that each PIP2 pool is likely to have a specific or restricted function within the cell. Thus, just beneath the NK-cell surface, we begin to see the murky world of membrane phospholipids coalesce into functionally meaningful patterns.