Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Apr 7.
Published in final edited form as: Mol Cell. 2016 Apr 7;62(1):21–33. doi: 10.1016/j.molcel.2016.03.009

Zinc-induced polymerization of killer-cell Ig-like receptor into filaments promotes its inhibitory function at cytotoxic immunological synapses

Santosh Kumar 1, Sumati Rajagopalan 1, Pabak Sarkar 2, David W Dorward 3, Mary E Peterson 1, Hsien-Shun Liao 4,8, Christelle Guillermier 5,6,7, Matthew L Steinhauser 5,6,7, Steven S Vogel 2, Eric O Long 1,*
PMCID: PMC4826557  NIHMSID: NIHMS768317  PMID: 27058785

SUMMARY

The inhibitory function of killer cell immunoglobulin-like receptors (KIR) that bind HLA-C and block activation of human natural killer (NK) cells is dependent on zinc. We report that zinc induced the assembly of soluble KIR into filamentous polymers, as detected by electron microscopy, which depolymerized after zinc chelation. Similar KIR filaments were isolated from lysates of cells treated with zinc, and membrane protrusions enriched in zinc were detected on whole cells by scanning electron microscopy and imaging mass spectrometry. Two independent mutations in the extracellular domain of KIR, away from the HLA-C binding site, impaired zinc-driven polymerization and inhibitory function. KIR filaments formed spontaneously, without addition of zinc, at functional inhibitory immunological synapses of NK cells with HLA-C+ cells. Adding to the recent paradigm of signal transduction through higher-order molecular assemblies, zinc-induced polymerization of inhibitory KIR represents an unusual mode of signaling by a receptor at the cell surface.

Keywords: Filament, Inhibition, Natural Killer cell, Signaling, Zinc

INTRODUCTION

Zinc is an essential trace element that contributes to many facets of biology. As an intrinsic component of proteins, it controls the catalytic activity of enzymes and the folding of proteins such as zinc fingers. In addition, zinc mediates protein assembly into dimers and oligomers. Examples include binding of tyrosine kinase Lck to CD4 (Huse et al., 1998) and storage of zinc-stabilized insulin hexamers in secretory vesicles (Li, 2014). Zinc acts also as a neurotransmitter, when released from synaptic vesicles in the hippocampus (Pan et al., 2011), and as a second messenger to regulate signal transduction in mast cells, dendritic cells, and T lymphocytes (Kitamura et al., 2006; Yamasaki et al., 2007; Yu et al., 2011). Zinc contributes to pathology by promoting amyloid fibril aggregation and deposition in the brain (Bush and Tanzi, 2002). Due to its high toxicity, zinc availability is tightly regulated through transporters and zinc-binding proteins. We reported earlier that zinc is required for the inhibitory function of an immunoreceptor that regulates the activity of cytotoxic innate lymphocytes called natural killer (NK) cells (Rajagopalan and Long, 1998; Rajagopalan et al., 1995).

NK cells are critical in the control of virus infections, in tumor surveillance, and regulation of adaptive immunity through direct cell contact and cytokine secretion (Iannello et al., 2016; Morvan and Lanier, 2015; Vivier et al., 2011; Waggoner et al., 2015). Their activity is tightly controlled by inhibitory receptors for major histocompatibility complex (MHC) class I (MHC-I) molecules, which are expressed on most cells. Human NK cells express killer cell immunoglobulin-like receptors (KIR) that bind to the MHC-I molecule HLA-C and exert powerful inhibition of NK cell activation (Long et al., 2013; Moretta et al., 1996). This inhibitory system has been exploited in the clinical setting of bone marrow transplantation: a mismatch between the specificity of inhibitory KIR in donor NK cells and HLA-C in transplant recipients favors NK cell activation, leading to graft-versus-leukemia activity and reduced graft-versus-host disease (Foley et al., 2014; Parham and McQueen, 2003).

Inhibitory KIRs block the polarization of lytic granules and degranulation at a very proximal step in the activation pathway for cellular cytotoxicity (Long et al., 2013). Accumulation of inhibitory KIR at NK–target cell immunological synapses is unusual in its independence of actin polymerization and dependence on zinc (Davis et al., 1999; Liu et al., 2012). The N-terminal zinc-binding motif (HExxH) of KIRs specific for HLA-C is required for their inhibitory function (Rajagopalan and Long, 1998).

To gain insight into the zinc dependence of KIR inhibitory function, we examined the biochemical properties of a purified soluble KIR protein. To our surprise, zinc was sufficient to induce assembly of KIR into filamentous polymers, which depolymerized upon zinc chelation. We provide evidence that this unique type of zinc-driven polymerization of a transmembrane receptor at the plasma membrane is required for the inhibitory function of KIR.

RESULTS

Zinc-induced polymerization of soluble KIR2DL1 into filaments

To investigate the effect of zinc on KIR2DL1 we purified the full ectodomain (Figure S1A, S1B), consisting of two Ig-like domains and a stem (amino acids 1–224), and measured its intrinsic Trp fluorescence spectra at different concentrations of ZnCl2. Trp fluorescence is sensitive to the hydrophobicity of its residing environment. KIR2DL1 has three Trp residues, at position 29, 188, and 207. In the absence of zinc, KIR2DL1 had maximum Trp fluorescence at ~ 348 nm (Figure 1A), suggesting that the Trp residues were partially exposed to solvent, consistent with our previous observation (Kumar et al., 2015). Treatment with ZnCl2 led to a shift in the wavelength of maximum fluorescence to ~ 339 nm, indicating a change in the environment of the Trp residue(s) in the receptor, and a concurrent increase in the fluorescence intensity. The transition to the zinc-induced state occurred in a narrow range of ZnCl2 concentrations, between 10 and 20 μM, consistent with a co-operative, binary switch-like transition (Figure 1B).

Figure 1. Zinc-induced polymerization of purified ectodomain of KIR2DL1 (1–224) into filaments.

Figure 1

Open in a new tab

(A) Intrinsic Trp fluorescence spectra. 1.5 μM soluble KIR2DL1 was treated with no zinc (blue), or 5 μM (red), 12.5 μM (green), 20 μM (dotted blue), and 50 μM (dotted red) ZnCl2 for 30 minutes. (B) Dependence of Trp fluorescence at 338 nm (I338 nm) on ZnCl2 concentration. Error bars represent standard deviations of the mean determined from three independent experiments. (C) Intrinsic Trp fluorescence spectra of KIR2DL1 in supernatant after centrifugation. 1.5 μM KIR2DL1 was treated with no zinc (blue), 50 μM ZnCl2 for 30 min (red), or 50 μM ZnCl2 for 30 min followed by 100 μM EDTA for 30 min (black). (D) TEM images after treatment of 1.5 μM KIR2DL1 with 50 μM ZnCl2 for 30 min. (E, F) TEM images after treatment of 1.5 μM KIR2DL1 with 50 μM ZnCl2 (E), or 50 μM ZnCl2 followed by 100 μM EDTA (F). (G) AFM image after treatment of 1.5 μM soluble KIR2DL1 with 50 μM ZnCl2 for 30 min. (H) Intrinsic Trp fluorescence spectra of native (dotted black) and urea-induced unfolded (dotted green) KIR2DL1. (I) Kinetics of refolding in the absence (black) and presence of 5 μM (red), 10 μM (solid blue), and 50 μM (dotted blue) ZnCl2. Dotted black and dotted green lines represent signals for native and unfolded receptor, respectively. (J) Intrinsic Trp fluorescence spectra of unfolded KIR2DL1 (dotted green) and of the end products of KIR2DL1 refolding in the absence of zinc (black), and in the presence of 5 μM (red), 10 μM (solid blue), and 50 μM (dotted blue) ZnCl2. In panels, H-J, the final concentration of KIR2DL1 was 1.5 μM. See also Figure S1.

The ZnCl2-treated receptor partitioned into the pellet after centrifugation at 15,000 g for 2 minutes, demonstrating aggregation of the receptor, which was reversed by EDTA treatment (Figure 1C). EDTA alone had no effect on the intrinsic Trp fluorescence spectrum of non-aggregated KIR2DL1 (Figure S1C). The morphology of zinc-induced aggregates of KIR2DL1 was examined by transmission electron microscopy (TEM). TEM images revealed filamentous polymers, often found in bundles (Figure 1D, 1E), which could account for their partitioning into the pellet upon centrifugation. Addition of EDTA disaggregated those filaments (Figure 1E, 1F). These results were confirmed by atomic force microscopy (AFM). AFM images revealed KIR2DL1 filaments of morphologies that were similar to those seen in TEM images (Figure 1G). We concluded that zinc induced polymerization of KIR2DL1 into higher-order filamentous assemblies.

Of the transition metals Cu2+, Ni2+, Co2+, and Fe2+, tested at 50 μM, Cu2+ alone induced a blue shift in the intrinsic Trp fluorescence spectrum of KIR2DL1 (Figure S1D). However, unlike Zn2+, Cu2+ did not induce an increase in the Trp fluorescence intensity. TEM images revealed Cu2+-induced aggregates of KIR2DL1 with a different morphology from that of Zn2+-induced filaments (Figure S1E).

To probe the initial events of zinc-induced KIR2DL1 polymerization into filaments, we examined the effect of ZnCl2 on receptor refolding from its unfolded state. Purified KIR2DL1 was unfolded in 6 M urea, and refolded by dilution into 0.6 M urea. Increased solvent exposure during unfolding in 6 M urea caused a shift in the wavelength of maximum fluorescence from ~ 348 nm (native protein) to ~ 358 nm (unfolded protein) (Figure 1H). The kinetics of refolding was monitored at 360 nm (I360 nm) (Figure 1I), where the largest difference in intensity between native and unfolded states occurred. Upon completion of the change in I360 nm, intrinsic Trp fluorescence spectra were measured (Figure 1J).

Unfolded KIR2DL1 refolded into its native structure within a few seconds, in the absence or presence of 5 μM ZnCl2 (Figure 1I), as judged by the native-like intrinsic Trp fluorescence spectrum after refolding (Figure 1J). In contrast, refolding in the presence of 10 μM or 50 μM ZnCl2 did not decrease the I360 nm value to that of the native receptor, but increased it within seconds to the value expected for zinc-induced filaments (Figure 1I). The intrinsic Trp fluorescence spectrum (Figure 1J) and TEM images (Figure S1F) of the end products showed that refolding from an unfolded state at 10 μM and 50 μM ZnCl2 led directly to filament formation. Whereas native KIR2DL1 transitioned to the ZnCl2-induced state between 10 and 20 μM ZnCl2 (Figure 1B), this transition occurred between 5 and 10 μM ZnCl2 with unfolded KIR2DL1 (Figure S1G). These results suggested that a partially folded state, formed during receptor refolding, is a preferred precursor for KIR2DL1 filament formation than the native receptor.

While Zn2+ and Cu2+ can promote amyloid fibrillation and deposition, such fibrils are resistant to metal chelation (Bush and Tanzi, 2002; Dong et al., 2014). In contrast, KIR2DL1 filaments were easily dissociated upon zinc chelation (Figure 1C, 1E, 1F). KIR2DL1 filaments showed weak binding to thioflavin T, a small amyloid-binding dye (LeVine, 1999), as compared to amyloid fibrils of the yeast prion protein Ure-2 (Sharma and Masison, 2011) (Figure S1H, S1I). Amyloid fibril formation leads to conformational change into cross-β structures (Chiti and Dobson, 2006), and is typically associated with an increase in β-sheet content. The two Ig domains of native KIR2DL1 are β-strand rich, with far-UV circular dichroism spectra showing a single peak at 212 nm (Fan et al., 1997; Kumar et al., 2015). After treatment with 50 μM ZnCl2 the far-UV circular dichroism spectrum of KIR2DL1 (1–224) was very similar, showing minimal change in ellipticity at 212 nm (Figure S1J, S1K). Thus, zinc-induced KIR2DL1 filamentous polymers are not typical of amyloid fibrils, and require zinc for their formation and maintenance.

Zinc-induced polymerization of KIR2DL1 into filaments in NK cells

While a soluble protein has no rotational constraint, the 2-dimensional space of the plasma membrane imposes constraints to transmembrane proteins that could hinder KIR2DL1 polymerization into filaments. To test it, we first used steady-state fluorescence anisotropy as a tool to measure KIR2DL1 self-association (Kumar et al., 2015). Homo-FRET between KIR2DL1 molecules fused to Venus fluorescence protein would result in decreased fluorescence anisotropy, indicative of self-association of the receptor (Nguyen et al., 2015). To measure thousands of cells at once, we performed anisotropy measurements in 96-well plate mode. In the absence of exogenous zinc, the anisotropy value in YTS cells expressing KIR2DL1 tagged with Venus at the cytosolic end (YTS–2DL1-Venus) remained relatively stable over time. After addition of ZnCl2, a drop in anisotropy was detected 10 min later, which remained almost stable over 1.5 h, and was reversed upon EDTA addition (Figure 2A). A decrease in anisotropy after addition of ZnCl2 was also observed with YTS cells expressing KIR2DL1 tagged with green fluorescence protein (GFP) at the extracellular N-terminus (YTS–GFP-2DL1), and was again completely reversed by chelation (Figure 2B). The large drop in anisotropy with YTS–GFP-2DL1 cells could be due to closer proximity of N-terminal GFP molecules, resulting in higher homo-FRET efficiency, as compared to the C-terminal Venus molecules. These results showed that ZnCl2 induced a sustained self-association of KIR2DL1 on YTS cells, which was reversible by chelation of ZnCl2.

Figure 2. Zinc-induced KIR2DL1 filament formation on NK cells.

Figure 2

Open in a new tab

(A, B) Steady-state anisotropy measurements in 96-well plate mode. 400 μM ZnCl2 and 1 mM EDTA were added at the indicated times to YTS–2DL1-Venus cells (A) or YTS–GFP-2DL1 cells (B). Error bars represent standard deviations of the mean determined from three (A) and seven (B) independent experiments. (C) Steady-state anisotropy measurements in imaging mode. YTS–2DL1-Venus cells were treated with 50 μM or 100 μM ZnCl2, as indicated, for 1.5 h. Anisotropy (upper panels) and fluorescence intensity (lower panels) images are shown. (D) Plot of median anisotropy against fluorescence intensity for every pixel in (C). (E–G) TEM images of KIR2DL1 immunoprecipitated from YTS–2DL1-Venus cells that were incubated with no zinc (E), 50 μM ZnCl2 (F), or 100 μM ZnCl2 (G) for 1.5 h.

Anisotropy measurements were also performed in imaging mode using two-photon excitation (Figure 2C). The median anisotropy value determined in the absence of exogenous zinc was ~ 0.3, which is similar to the value of a monomeric, membrane-tethered GFP (Kumar et al., 2015), suggesting that KIR2DL1 was monomeric. Addition of ZnCl2 at 50 and 100 μM led to a drop in anisotropy (Figure 2C), which was independent of fluorescence intensity, as shown by plotting the anisotropy value of each pixel as a function of fluorescence intensity (Figure 2D). These results showed that self-association of KIR2DL1 induced by ZnCl2 for 1.5 h occurred even at low receptor density and was not due to receptor crowding.

To examine the aggregation state and morphology of self-associated KIR2DL1 on NK cells, immunoprecipitated KIR2DL1-Venus was imaged by TEM. In the absence of zinc, no discernable structure appeared within the background (Figure 2E). In contrast, filamentous polymers were observed in immunoprecipitates from YTS–2DL1-Venus cells treated with 50 μM and 100 μM ZnCl2 (Figure 2F, 2G). Although we cannot exclude post-lysis polymerization, these results suggest, consistent with anisotropy data, that ZnCl2 induced polymerization of KIR2DL1 into filaments at the plasma membrane of NK cells, with a morphology similar to that formed in vitro with soluble KIR2DL1.

The KIR2DL1 N-terminal HExxH motif is required for zinc-induced polymerization into filaments

Two N-terminal His residues at position 1 and 5 in KIR2DL1 are part of a classical HExxH zinc-binding motif. Mutation of both His residues to Ala (mutant H1,5A) in KIR2DL1 resulted in defective inhibitory function of cloned, primary NK cells (Rajagopalan and Long, 1998) and of YTS–2DL1-H1,5A cells (Figure S2). The defect was at the level of inhibitory signaling and not ligand binding, as a soluble KIR2DL1-H1,5A Fc fusion protein bound to HLA-C+ cells as well as wild type KIR2DL1 (Rajagopalan and Long, 1998). Furthermore, fusion of KIR2DL1-H1,5A with the activating, immunoreceptor tyrosine-based activation motif (ITAM)-bearing FcR γ-chain in place of the immunoreceptor tyrosine-based inhibition motif (ITIM)-bearing cytosolic tail resulted in a chimeric receptor that induced killing of HLA-C+ target cells (Rajagopalan and Long, 1998). To test if the HExxH motif was also required for zinc-induced polymerization, we purified a soluble form of the ectodomain (amino acids 1–224) of KIR2DL1-H1,5A. This H1,5A mutant underwent a zinc-dependent change in intrinsic Trp fluorescence (Figure 3A). However, the transition from the native to the zinc-induced state was not as sharp as with wild type KIR2DL1 (Figure 3B). TEM images of aggregates of KIR2DL1-H1,5A induced by 50 μM ZnCl2 were strikingly different from wild-type KIR2DL1, displaying large, amorphous aggregates rather than filaments (Figure 3C). Thus, the HExxH motif required for inhibitory signaling by KIR2DL1 was also required for zinc-mediated polymerization into linear filaments.

Figure 3. The N-terminal HExxH motif is required for zinc-induced polymerization of KIR2DL1 into filaments.

Figure 3

Open in a new tab

(A) Intrinsic Trp fluorescence spectra of 1.5 μM soluble KIR2DL1-H1,5A treated with no zinc (blue) or 5 μM (red), 12.5 μM (green), 20 μM (dotted blue), 30 μM (black), and 50 μM (dotted red) ZnCl2 for 30 min. (B) Dependence of Trp fluorescence at 338 nm (I338 nm) on ZnCl2 concentration for wild type KIR2DL1 (circles) and KIR2DL1-H1,5A (squares). Error bars represent standard deviations of the mean determined from three independent experiments. (C) TEM images of soluble KIR2DL1-H1,5A treated with 50 μM ZnCl2 for 30 min. (D) Anisotropy measurements on YTS–2DL1-H1,5A-Venus cells treated as in Figure 2. Error bars represent standard deviations of mean determined from three independent experiments. (E, F) TEM images of KIR2DL1-H1,5A immunoprecipitated from YTS–2DL1-H1,5A-Venus cells treated for 1.5 h in the absence (E) or presence (F) of 200 μM ZnCl2. See also Figure S2.

We next examined the H1,5A mutant on YTS–2DL1-H1,5A cells treated with exogenous ZnCl2. As seen with wild-type KIR2DL1-Venus, the anisotropy value decreased after addition of exogenous ZnCl2, and was restored after chelation by EDTA (Figure 3D). The zinc-induced decrease in anisotropy for the H1,5A mutant was greater than that of wild-type KIR2DL1 (Figure 2A, 3D). Thus, as seen with the purified receptor in solution, the type of polymer formed upon zinc-induced self-association of KIR2DL1-H1,5A on NK cells differed from the zinc-induced polymers of the wild-type receptor. Moreover, similar to the large amorphous aggregates formed in vitro, the H1,5A mutant on cells formed dense aggregates in the presence of ZnCl2, yielding a larger homo-FRET signal. No filaments were detected by TEM in immunoprecipitates of KIR2DL1 from YTS–2DL1-H1,5A-Venus cells treated with ZnCl2 (Figure 3E, 3F). Thus, the two N-terminal His were not required for ZnCl2–induced KIR2DL1 aggregation, but were essential for polymerization into linear filaments.

The stem region in the extracellular domain of KIR2DL1 is required for zinc-induced polymerization in NK cells and for inhibitory function

The KIR2DL1 Ig domains are connected to the transmembrane region by a Pro-rich stem. This stem was not visible in the crystal structure of a KIR2DL1–HLA-Cw4 complex, indicating that it is disordered and not required for binding to HLA-C (Fan et al., 2001). Removal of the stem region resulted in a KIR2DL1 mutant (KIR2DL1-SD) that lost its inhibitory function (Figure S3).

We purified a soluble, truncated KIR2DL1 (1–200) lacking the stem to test zinc-induced filament formation. A blue shift in the intrinsic Trp fluorescence spectra occurred between ~ 6 and 30 μM ZnCl2 (Figure 4A, 4B), indicating that ZnCl2 induced a change in the environment of the Trp residues. Unlike wild-type KIR2DL1, there was very little change in Trp fluorescence intensity (Figure 4A). This could be due to the lack of Trp residue 207 or to a different type of change in the vicinity of Trp residues. TEM images of ZnCl2-treated KIR2DL1 (1–200) revealed filamentous aggregates, somewhat similar to those observed with wild-type KIR2DL1 (Figure 4C). We concluded that the stem region was not essential for ZnCl2-induced polymerization in vitro. However, the stem could contribute to the final arrangement of KIR2DL1 linear polymers.

Figure 4. The stem region of KIR2DL1 is required for zinc-induced polymerization into filaments on cells and for inhibitory function.

Figure 4

Open in a new tab

(A) Intrinsic Trp fluorescence spectra of 1.5 μM soluble KIR2DL1-SD treated with no zinc (blue), or 5 μM (red), 12.5 μM (green), 20 μM (dotted blue), and 50 μM (dotted red) ZnCl2. (B) Dependence of the wavelength of maximum fluorescence (λmax) on ZnCl2 concentration. (C) TEM images of soluble KIR2DL1-SD treated with 50 μM ZnCl2 for 30 min. (D, E) TEM images of KIR2DL1-SD immunoprecipitated from YTS–2DL1-SD-Venus cells treated in the absence (D) or presence (E) of 100 μM ZnCl2 for 1.5 h. (F) Scattering intensity of KIR2DL1 immunoprecipitated from YTS–2DL1-Venus (WT) and YTS–2DL1-SD-Venus cells treated in the absence or presence of 100 μM ZnCl2, as indicated, for 1.5 h. The value of scattering intensity obtained in the absence of exogenous zinc was set to 1. Error bars represent standard errors of the mean determined from four independent experiments. **p<0.005. Statistical analysis was done by non-parametric t test using GraphPad (prism). (G) Staining with anti-HLA-C antibody F4/326 of 221–HLA-Cw4–ICP47 cells that had been incubated with increasing amounts of an HLA-Cw4 specific peptide for 16 h at 26°C. (H) Lysis of peptide-loaded 221–HLA-Cw4–ICP47 cells by YTS–2DL1-Venus (circles), YTS–2DL1-H1,5A-Venus (triangles), and YTS–2DL1-SD-Venus (squares) cells. Lysis of target cells in the absence of exogenous peptide was set to 1. Error bars represent SD from three independent experiments. See also Figure S3.

We next tested the role of the stem region in filament formation at the plasma membrane. TEM images of stem-deleted KIR2DL1 immunoprecipitated from YTS–2DL1-SD-Venus cells that had been treated with exogenous ZnCl2 did not reveal any filament or ordered structure (Figure 4D, 4E). To quantitate the presence of polymers in immunoprecipitated samples we performed light scattering measurements. ZnCl2 treatment of YTS–2DL1-Venus cells resulted in increased scattering intensity of KIR2DL1 immunoprecipitates, which was not observed after ZnCl2 treatment of YTS–2DL1-SD-Venus cells (Figure 4F). These results confirmed the information obtained by TEM and showed that the stem region, which is required for the inhibitory function of KIR2DL1, is also essential for ZnCl2-induced polymerization into filaments on cells.

We used a quantitative assay for inhibition (Kumar et al., 2015) to test the function of stem-deleted KIR2DL1 in YTS–2DL1-SD-Venus cells. 721.221 cells transfected with HLA-Cw4 and the viral transporter for antigen presentation (TAP) inhibitor ICP47 (221–HLA-Cw4–ICP47 cells), which blocks loading of endogenous peptides, were loaded with increasing amounts of an HLA-Cw4-specific peptide compatible with KIR2DL1 binding (Rajagopalan and Long, 1997). Expression of stable HLA-Cw4 at the cell surface was monitored by flow cytometry (Figure 4G). YTS cells expressing KIR2DL1 were tested for their ability to kill 221–HLA-Cw4–ICP47 cells. Whereas inhibition of cytotoxic activity of YTS–2DL1-Venus cells was commensurate with the amount of HLA-Cw4 at the surface of 221 target cells, inhibition of YTS–2DL1-SD-Venus and YTS–2DL1-H1,5A-Venus cells was completely impaired (Figure 4H). Given that binding of HLA-Cw4 tetramers to YTS–2DL1-Venus cells was not affected by the H1,5A or by the stem deletion mutations (Figure S4), the defect in inhibitory function of KIR2DL1-H1,5A and KIR2DL1-SD is most likely due to an impaired signaling function caused by the lack of zinc-induced polymerization. These results also suggested that the stem may be required to allow KIR2DL1 to adopt an orientation and flexibility that is compatible with an ordered zinc-induced polymerization at the plasma membrane.

Formation of zinc-induced, KIR2DL1-dependent membrane protrusions on intact cells

To exclude the possibility of post-lysis polymerization of KIR2DL1 into linear filaments, we examined intact, fixed cells by scanning electron microscopy (SEM). We purposely chose relatively high ZnCl2 concentrations (50 to 200 μM) and a long incubation time (2 h) in order to allow growth of KIR2DL1 polymers and improve their detection. SEM images revealed long filopodia-like membrane protrusions at the surface of ZnCl2-treated cells that expressed KIR2DL1 (Figure 5A). These filopodia-like protrusions were not observed on untreated KIR2DL1-expressing cells (Figure 5A), or on untransfected cells treated with 200 μM ZnCl2 (Figure 5B). Zinc-induced membrane protrusions were also observed after transient expression of KIR2DL1 on HEK293T cells (Figure 5B). Bundling of parallel filopodia-like protrusions occurred (Figure 5A, 5B), reminiscent of zinc-induced filamentous polymer bundles of soluble KIR2DL1 in vitro. We concluded that the combination of transmembrane KIR2DL1 and ZnCl2 promoted the growth of membrane protrusions on intact cells.

Figure 5. High-resolution imaging of whole cells after KIR2DL1 expression and treatment with ZnCl2.

Figure 5

Open in a new tab

(A) SEM images of HEK293T cells stably transfected with KIR2DL1 (293T–2DL1) that had been treated with no ZnCl2 or 200 μM ZnCl2 for 2 h. The right-most panel is a higher magnification of membrane protrusions. Scale bars represent 5 μm in the first two panels, and 500 nm in the right panel. (B) SEM images of HEK293T cells that had been either mock-transfected or transfected with KIR2DL1, as indicated, and treated with 200 μM ZnCl2. The right-most panel is an enlargement of the middle panel. Scale bars represent 5 μm in the first two panels, and 1 μm in the right panel. (C) NanoSIMS images of HEK293T cells that were either untransfected or stably transfected with KIR2DL1 (293T–2DL1), and treated with the indicated ZnCl2 concentrations for 2 h. Mass images, as determined by CN− emission-based pixel intensities, are shown. Scale bars represent 5 μm. (D) Higher resolution NanoSIMS images of membrane protrusions on 293T–2DL1 cells. CN− (left-most panel) and ZnO− (middle and right-most panels) images were acquired simultaneously. In the right-most panel, ZnO− counts within the indicated area (yellow outline) are pseudocolored red. Scale bars represent 2 μm. (E) The mean ZnO− emission values obtained from the filopodia-like protrusions on 293T–2DL1 cells, treated with 50 μM and 200 μM ZnCl2, were normalized relative to those obtained from adjacent regions of the silicon support (background). Error bars represent standard error of the mean obtained from analysis of multiple cells for each of the ZnCl2 concentrations. * p<0.05. GraphPad (prism) was used to perform the t-test. (F) ZnO−/CN− ratio of cell edges and protrusions on 293T–2DL1 cells that were treated with 50 μM and 200 μM ZnCl2. The ZnO−/CN− ratios were normalized relative to randomly selected areas in the cell body. Error bars represent standard error of the mean obtained from analysis of multiple cells for each of the ZnCl2 concentrations. *p<0.05. A non-parametric Mann-Whitney test was used to compare the values for protrusions to those for the corresponding cell edge.

To obtain evidence that zinc-induced KIR2DL1-dependent filopodia-like protrusions on intact cells contained zinc-induced KIR2DL1 polymers, we turned to high resolution imaging mass spectrometry, an advanced technology that achieves high-resolution (~50 nm) scanning of samples, combined with mass spectrometry analysis of stable isotopes (Steinhauser and Lechene, 2013). CN− (i.e. carbon-nitrogen) ions are emitted from nitrogenous cellular constituents (e.g. proteins, nucleic acids), revealing subcellular structures, and, in this case, demonstrating membrane protrusions emanating from the surface of HEK293T cells expressing KIR2DL1, similar to SEM images (Figure 5C). ZnO− ions were detected simultaneously as a measure of zinc composition. As expected for direct measurement of a trace element, the counts of ZnO− were low and therefore difficult to visualize (Figure 5D). Nevertheless, ZnO− measurements from the protrusions were reproducibly, and significantly, higher than the emission from the surrounding silicon support (Figure 5E), and the emission of ZnO− from the cell body, including cell margins (Figure 5F). These data support the concept that formation of filopodia-like protrusions is not only dependent upon zinc, but that the protrusions themselves contain zinc.

Zinc-induced polymerization is not sufficient for tyrosine phosphorylation of KIR2DL1 and causes loss of antibody and HLA-C binding

As receptor clustering is often associated with phosphorylation of intracellular tyrosine residues, we examined KIR2DL1 phosphorylation in YTS–2DL1-Venus cells treated with ZnCl2. Parallel immunoprecipitations with Abs to GFP and to SHP-1, followed by gel electrophoresis and immunoblotting for phospho-Tyr, revealed that zinc treatment did not result in KIR2DL1 phosphorylation (Figure 6A, 6B). To test if ITIM sequences in zinc-induced polymers were accessible to tyrosine kinases, we treated YTS–2DL1-Venus cells with a limiting amount of pervanadate, an inhibitor of tyrosine phosphatases. As expected, pervanadate alone induced Tyr phosphorylation of KIR2DL1 (Figure 6A, 6B). As tyrosine phosphatase SHP-1 is recruited selectively to phosphorylated ITIMs (Burshtyn et al., 1996), SHP-1 immunoprecipitates demonstrated direct phosphorylation of ITIM (Figure 6B). Furthermore, prior treatment of cells with ZnCl2 did not prevent tyrosine phosphorylation of KIR2DL1-Venus induced by pervanadate, but, in fact, resulted in enhanced phosphorylation (Figure 6A, 6B). Similar results were obtained with KIR2DL1 in NKL–2DL1-SBP cells, which had been pulled down with streptavidin (Figure 6C). ZnCl2 had no obvious effect on the basal, total amount of tyrosine phosphorylation in YTS–2DL1-Venus cells (Figure S4A) and NKL–2DL1-SBP cells (Figure S4B). However, prior treatment of cells with ZnCl2 resulted in a small increase of the total cellular tyrosine phosphorylation induced by pervanadate (Figure S4A, S4B). Therefore, the small ZnCl2-dependent enhancement of KIR2DL1 phosphorylation induced by pervanadate (Figure 6A, 6B, 6C) may not be due to KIR2DL1 polymerization but may simply reflect the enhancement of global tyrosine phosphorylation (Figure S4). We concluded that ZnCl2 on its own did not induce tyrosine phosphorylation of KIR2DL1.

Figure 6. Zinc-induced polymerization is not sufficient for KIR2DL1 tyrosine phosphorylation and causes loss of antibody and HLA-C binding.

Figure 6

Open in a new tab

(A–C) YTS–2DL1-Venus and NKL–2DL1-SBP cells were treated sequentially with 200 μM ZnCl2 for 1.5 h, and pervanadate for 5 min. Cells receiving a single, or no treatment are indicated below each lane. Immunoprecipitated samples were immunoblotted for p-Tyr. (A, B) KIR2DL1-Venus was immunoprecipitated with an anti-GFP nanobody (A) or an Ab to SHP-1 (B) from YTS–2DL1-Venus cells treated with ZnCl2 and 200 μM pervanadate, as indicated. The position of KIR2DL1-Venus and SHP-1 in panel B is indicated by * and **, respectively. (C) KIR2DL1-SBP was pulled-down with streptavidin beads from NKL–2DL1-SBP cells treated with ZnCl2 and 800 μM pervanadate, as indicated. (D, E) YTS–2DL1-Venus cells were stained with anti-KIR2DL1 mAb (clone EB6) coupled with allophycocyanin (APC), after treatment with no zinc (circles), 100 μM ZnCl2 (squares), or 200 μM ZnCl2 (triangles) for the indicated time. (E) Venus fluorescence was acquired simultaneously on the same samples shown in (D). The data shown in (D) and (E) are representative of two independent experiments. (F) YTS–GFP-2DL1 cells were treated with ZnCl2 for 1.5 h and stained with anti-KIR2DL1 mAb (clone 143211) (circles) or anti-GFP Ab (squares). Error bars represent standard deviations of the mean determined from three independent experiments. (G) YTS–2DL1-Venus cells were treated with ZnCl2 for 1.5 h and stained with Abs to KIR2DL1 (clone EB6) (circles), receptor 2B4 (triangles), CD28 (inverted triangles), and MHC class I (diamonds). Abs were directly coupled with phycoerythrin. The error bars represent standard deviations of the mean determined from three independent experiments. (H) YTS–2DL1-Venus cells were stained with anti-KIR2DL1 mAb (clone 143211) either before (squares) or after (circles) treatment with ZnCl2 for 1.5 h. An anti-IgG antibody was used to detect the primary Ab. (I) YTS–2DL1-Venus cells were stained with anti-KIR2DL1 mAb (EB6-PE) (squares), and with HLA-Cw4 tetramer coupled to Alexa 647 (triangles) after treatment with the indicated concentrations of ZnCl2 for 2 h. Venus fluorescence was acquired at the same time (circles). (J) NKL–2DL1-SBP cells were treated and stained as in (I) with the addition of an anti-CD94-FITC antibody (circles). See also Figure S4.

We noticed that binding of a mAb to KIR2DL1 was diminished after treatment of YTS–2DL1-Venus cells with ZnCl2 (Figure 6D). The loss of mAb binding on cells treated with 100 μM or 200 μM ZnCl2 progressed slowly, over several hours, while the Venus signal from the same cells remained steady (Figure 6E). As internalization of KIR2DL1 into acidic compartments would be expected to reduce Venus fluorescence through quenching or degradation, these results suggested that the mAb binding was lost as a result of polymerization. Zinc treatment of YTS cells expressing KIR2DL1 tagged with GFP at the extracellular N-terminus (YTS–GFP-2DL1) resulted also in the loss of anti-KIR2DL1 mAb binding, but not binding of an anti-GFP mAb (Figure 6F), demonstrating that KIR2DL1 was not lost by internalization. Furthermore, zinc-induced decrease of Ab binding to KIR2DL1 was selective, as staining for other receptors was not affected (Figure 6G). Binding of the anti-KIR2DL1 mAb to YTS–2DL1-Venus cells prior to addition of ZnCl2 prevented the loss of mAb caused by ZnCl2 (Figure 6H). Taken together, these results demonstrated a loss of Ab epitopes on KIR2DL1, as observed with two different Abs, which could be due to a conformational change in KIR2DL1 or to the masking of epitopes in the core of zinc-induced polymers.

Binding of soluble HLA-Cw4 tetramer to YTS–2DL1-Venus cells and to NKL–2DL1-SBP was also lost after zinc treatment of the cells (Figure 6I, 6J). The relative drop of mAb and of HLA-C ligand binding at different ZnCl2 concentrations had the same slope. We concluded that zinc-induced KIR2DL1 polymerization on cells resulted in reduced binding of KIR2DL1 with its MHC-I ligand.

KIR2DL1 filamentous polymers at functional NK–target cell inhibitory synapses

The impaired ability of mutants KIR2DL1-H1,5A and KIR2DL1-SD to form zinc-induced filamentous polymers on NK cells and to signal at inhibitory NK–target cell synapses suggested that zinc-induced polymerization of wild-type KIR2DL1 is required for its inhibitory function. If so, one would predict formation of filamentous polymers of KIR2DL1 at inhibitory synapses between KIR2DL1+ NK cells and target cells expressing a cognate HLA-C ligand, in the absence of added exogenous zinc. This was tested by mixing YTS–2DL1-Venus cells with 721.221 target cells expressing HLA-Cw3 (an allotype that does not bind KIR2DL1) or HLA-Cw15 (an allotype that binds KIR2DL1). TEM images of immunoprecipitated KIR2DL1-Venus revealed filamentous polymers in cells that had formed inhibitory, but not activating synapses (Figure 7A). AFM images confirmed these results (Figure 7B). We concluded that filamentous polymers of KIR2DL1 formed rapidly and selectively at functional inhibitory NK–target cell synapses, and in the absence of addition of exogenous zinc.

Figure 7. KIR2DL1 filamentous polymers at functional inhibitory NK target cell synapses.

Figure 7

Open in a new tab

(A, B) TEM images (A), and AFM images (B) in topography mode, of KIR2DL1 immunoprecipitated from YTS–2DL1-Venus cells that had been mixed with activating 221–HLA-Cw3 and inhibitory 221–HLA-Cw15 target cells. Scale bars are 100 nm (A) and 330 nm (B). (C) TEM images of KIR2DL1 pulled-down with streptavidin beads from NKL–2DL1-SBP cells that had been mixed with activating 221 and inhibitory 221–HLA-Cw15 target cells. The top rows show images from one experiment, and the rest are images from a separate experiment. The bottom row shows images of KIR2DL1 pull-down samples treated with 25 μM TPEN for 30 min or 100 μM EDTA for 30 min, as indicated. Scale bars represent 100 nm.

Next, we examined filament formation at synapses formed between NKL–2DL1-SBP cells and 221 target cells. TEM images of KIR2DL1-SBP pulled down with streptavidin and eluted with biotin showed filaments in samples from inhibitory synapses with 221–HLA-Cw15 cells (Figure 7C). Samples from activating synapses with untransfected 221 cells did not show such distinct structures (Figure 7C). To test if the KIR2DL1 filaments that formed at NK cell inhibitory synapses were similar to those obtained with soluble KIR2DL1 in the presence of ZnCl2, the samples pulled-down with streptavidin from inhibitory synapses were treated with EDTA and with the zinc-specific chelator TPEN. No filamentous polymers were detected after zinc chelation (Figure 7C), demonstrating that KIR2DL1 filament formation at functional NK cell inhibitory synapses was zinc-dependent.

DISCUSSION

Our study provides evidence for a zinc-induced polymerization of receptor (ZiPR) mechanism, which is required for the inhibitory function of a receptor that controls activation of human NK cells. In the physiological context of inhibitory NK–target cell immunological synapses, binding of KIR2DL1 to HLA-C on target cells resulted in its polymerization into filaments that were sensitive to zinc chelation. This mechanism has several unusual features: the extracellular domain of a transmembrane receptor polymerizes into filaments, polymerization is zinc-dependent, and zinc-induced polymers at the cell surface contribute to inhibitory signaling by the receptor.

Two KIR2DL1 mutant variants that were compromised in formation of filamentous polymers on cells treated with zinc had impaired function, despite HLA-C ligand binding. The mutations were at each end of the KIR2DL1 extracellular domain: an N-terminal zinc-binding motif and a C-terminal stem that separates the Ig domains from the transmembrane region. Those two terminal segments of KIR2DL1 showed no electron density in the crystal structure, indicating a disordered structure, and do not contribute any contact with HLA-C (Fan et al., 2001). KIR2DL1 mutated at the two N-terminal His residues (H1,5A) formed large, non-linear aggregates in the presence of zinc in solution. This N-terminal zinc-binding motif therefore controls the formation of an organized, linear polymer. Strong self-association of KIR2DL1-H1,5A on zinc-treated cells did not result in linear polymers detectable by TEM. KIR2DL1 lacking the stem did form filaments in the presence of zinc in solution but not when expressed as a transmembrane receptor on cells. The role of the Pro-rich stem in promoting polymerization at the plasma membrane, but not in solution, may be to release spatial constraints imposed by insertion across a 2-dimensional lipid bilayer.

Zinc promoted polymerization of purified, soluble KIR2DL1 into filaments, which was reversed by chelation of zinc. There is an exquisite specificity for Zn2+ in KIR2DL1 polymerization into filaments. Among other metals tested, only Cu2+ induced detectable changes, which resulted in polymerization into structures with a different morphology. This degree of selectivity is analogous to the formation of β2-microglobulin oligomers in the presence of Cu2+, which are reversible by chelation prior to their progression into irreversible β2-microglobulin amyloid fibrils (Calabrese et al., 2008). Zn2+, in contrast, induced β2-microglobulin polymerization into amorphous aggregates (Dong et al., 2014).

On cells treated with zinc, KIR2DL1 self-associated, as determined by anisotropy measurements, and polymerized into filaments visible by TEM and AFM after immunoprecipitation. Zinc promoted the formation of filopodia-like membrane protrusions, as shown by SEM on intact cells expressing KIR2DL1. Furthermore, similar membrane protrusions were detected by high-resolution imaging mass spectrometry on whole cell scans, and were enriched for zinc, as measured by emission of secondary zinc oxide (64Zn16O−) ions. How KIR filament formation on cells links to the tubulation responsible for the membrane protrusions is not known. Tight parallel alignment of some of these protrusions suggest that KIR2DL1 polymers may lie longitudinally, given that bundles of parallel filaments are the predominant form of zinc-treated, soluble KIR2DL1. In any case, KIR2DL1 filaments that form at functional inhibitory synapses within a few minutes may well have a different impact on membrane topology.

Zinc-induced polymerization did not result in tyrosine phosphorylation of the cytoplasmic ITIM sequences in KIR2DL1. Furthermore, binding of antibodies and of soluble HLA-C tetramers to KIR2DL1 on cells that had been treated with zinc was greatly reduced, despite retention of KIR2DL1 at the plasma membrane. Therefore, the role of KIR2DL1 polymers is not to increase valency for binding to HLA-C on target cells. As binding of KIR2DL1 to HLA-C on target cells is required for phosphorylation, it has to precede, or coincide with, zinc-induced polymerization. There is potential for signal amplification through serial engagement of HLA-C molecules, each of which could seed one or more polymers.

In contrast to the clustering and movement of receptors at activating immunological synapses, KIR accumulation at NK cell inhibitory synapses is independent of ATP and of actin reorganization (Davis et al., 1999; Liu et al., 2012). This unusual actin-independent clustering and signaling may explain the ability of KIR to block the actin-dependent recruitment and phosphorylation of activation receptors (Long, 2008; Stebbins et al., 2003; Watzl and Long, 2003). As KIR accumulation at inhibitory synapses is dependent on zinc (Davis et al., 1999), our results suggest that zinc-induced polymerization contributes to the actin-independent clustering of KIR. Using a photoswitchable fluorescent protein, a recent study showed that KIR2DL1 freely diffuses before reaching the inhibitory synapse, where it becomes trapped. Modeling and biophysical measurements suggested that KIR2DL1 is not retained through receptor–ligand binding (Pageon et al., 2013). Perhaps zinc-induced polymers contribute to KIR2DL1 retention at the synapse by limiting their diffusion.

The polymerization of KIR2DL1 into filaments at functional inhibitory synapses did not require addition of exogenous free zinc, underscoring the physiological importance of zinc transport and availability during NK–target cell contacts. Although the concentration of zinc in human serum is about 20 μM, very little of it is freely available (Bozym et al., 2010). Therefore, additional regulation of inhibitory KIR2DL1 function may occur through zinc delivery at NK cell immunological synapses.

While prion-like fibrils have long been associated with pathogenesis, recent work has identified fibrillar polymers of innate immune receptors and signaling components that form in the cytosol in response to infection and transmit signals for inflammation, immunity, and death pathways (Cai and Chen, 2014; Hou et al., 2011; Kagan et al., 2014; Li et al., 2012; Siegel et al., 1998). In contrast to these large, irreversible amyloid fibrils in the cytosol, polymers of inhibitory KIR at immunological synapses form at the cell surface, and dissociate in the absence of zinc. The discovery of zinc-dependent KIR filaments at inhibitory NK cell immunological synapses expands on the recent paradigm of functional higher-order assemblies of molecules that can regulate the amplitude, time and location of signals (Balagopalan et al., 2015; Cai and Chen, 2014; Wu, 2013). Finally, the ZiPR mechanism may have relevance beyond inhibitory immunological synapses to receptors in other systems, such as synaptic plasticity in the cerebral cortex, which is regulated by release of zinc from synaptic vesicles (Pan et al., 2011).

EXPERIMENTAL PROCEDURES

Recombinant, soluble KIR2DL1 was expressed in E. coli and purified as described (Kumar et al., 2015). Trp fluorescence was determined by excitation at 295 nm and measuring its emission at 310–410 nm. TEM of negatively stained samples was performed using methylamine tungstate. Scanning electron microscopy was performed on fixed cells adsorbed onto 0.05% poly-L-lysine silicon chips. Images were captured on a SU8000 scanning electron microscope (Hitachi High-Technologies, Dallas, TX). For AFM, samples were adhered onto freshly cleaved mica, rinsed with water and dried before imaging on a MultiMode AFM instrument (Bruker). Kinetics of KIR2DL1 refolding was determined by measuring Trp fluorescence at 360 nm. YTS cells expressing KIR2DL1-Venus have been described (Kumar et al., 2015). KIR2DL1 with two N-terminal His mutated to Ala (H1,5A) and KIR2DL1 lacking the stem region (200–221), each tagged with Venus, were cloned into Lentivector pCDH-EF1-MCS-T2A-Puro (System Biosciences). Transduced YTS cells were selected in 1 μg/ml puromycin. NKL cells transduced with the same Lentivector carrying KIR2DL1 tagged with a C-terminal streptavidin binding peptide were selected in 0.3 μg/ml puromycin. Anisotropy measurements in imaging mode were performed on cells plated onto a poly-L-lysine coated glass-bottom dish and imaged with a DCS-120 Confocal Scanning FLIM System (Becker & Hickl GmbH). Anisotropy measurements in plate mode were performed using cells in 96-well plates with a PheraStar FS microplate reader (BMG Labtech) using a fluorescence polarization module FP-490-540-540. Light scattering intensity measurements were performed by exciting samples at 405 nm and measuring emission at the same wavelength on a Fluoromax-3 spectrofluorometer (Jobin Yvon). The inhibitory function of KIR2DL1 and its mutants expressed in YTS cells was tested using HLA-C peptide loaded 721.221 cells coexpressing HLA-C*04:01 and TAP inhibitor ICP47 (221–Cw4–ICP47 cells), as described (Kumar et al., 2015). For high resolution imaging mass spectrometry, secondary 12C14N− and 64Zn16O− ions from cells, generated by sputtering the sample surface with a primary cesium ion beam, were measured simultaneously in a NanoSIMS 50L instrument (Cameca, France). KIR2DL1 filaments were isolated from YTS–2DL1-Venus cells and NKL–2DL1-SBP cells by immunoprecipitation with anti-GFP antibody and by pull-down with streptavidin beads, respectively.

Supplementary Material

supplement

Acknowledgments

We thank David Narum and Karine Reiter for assistance with circular dichroism, and Shailesh Kumar for amyloid fibrils of Ure-2. HLA-C*0401 tetramer was made at the NIH tetramer facility. The authors report no conflict of interest. This work was supported by the Intramural Research Programs at the National Institute of Allergy and Infectious Diseases (E.O.L.), National Institute on Alcohol Abuse and Alcoholism (S.S.V.), and National Institute of Biomedical Imaging and Bioengineering (H.-S.L.), National Institutes of Health, and by NIH grants K08DK090147 and RO3DK106477 (M.L.S.).

Footnotes

SUPPLEMENTAL INFORMATION

Supplemental information includes detailed Experimental Procedures, three Figures and one Table.

AUTHOR CONTRIBUTIONS

Conceptualization: S.K., and E.O.L.; Methodology/Design: S.K., S.R., P.S., C.G., M.L.S., S.S.V., and E.O.L.; Investigation: S.K., S.R., P.S., D.W.D., M.E.P., H.-S.L., C.G., and M.L.S.; Writing – Original Draft: S.K., S.R., and E.O.L.; Writing – Review & Editing: S.K., S.R., M.L.S., S.S.V., and E.O.L.; Supervision and Funding Acquisition: M.L.S., S.S.V., and E.O.L.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Balagopalan L, Kortum RL, Coussens NP, Barr VA, Samelson LE. The linker for activation of T cells (LAT) signaling hub: from signaling complexes to microclusters. J Biol Chem. 2015;290:26422–26429. doi: 10.1074/jbc.R115.665869. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bozym RA, Chimienti F, Giblin LJ, Gross GW, Korichneva I, Li Y, Libert S, Maret W, Parviz M, Frederickson CJ, Thompson RB. Free zinc ions outside a narrow concentration range are toxic to a variety of cells in vitro. Exp Biol Med. 2010;235:741–750. doi: 10.1258/ebm.2010.009258. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Burshtyn DN, Scharenberg AM, Wagtmann N, Rajagopalan S, Berrada K, Yi T, Kinet JP, Long EO. Recruitment of tyrosine phosphatase HCP (SHP-1) by the killer cell inhibitor receptor. Immunity. 1996;4:77–85. doi: 10.1016/s1074-7613(00)80300-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bush AI, Tanzi RE. The galvanization of beta-amyloid in Alzheimer’s disease. Proc Natl Acad Sci U S A. 2002;99:7317–7319. doi: 10.1073/pnas.122249699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cai X, Chen ZJ. Prion-like polymerization as a signaling mechanism. Trends Immunol. 2014;35:622–630. doi: 10.1016/j.it.2014.10.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Calabrese MF, Eakin CM, Wang JM, Miranker AD. A regulatable switch mediates self-association in an immunoglobulin fold. Nat Struct Mol Biol. 2008;15:965–971. doi: 10.1038/nsmb.1483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chiti F, Dobson CM. Protein misfolding, functional amyloid, and human disease. Annu Rev Biochem. 2006;75:333–366. doi: 10.1146/annurev.biochem.75.101304.123901. [DOI] [PubMed] [Google Scholar]
  8. Davis DM, Chiu I, Fassett M, Cohen GB, Mandelboim O, Strominger JL. The human natural killer cell immune synapse. Proc Natl Acad Sci U S A. 1999;96:15062–15067. doi: 10.1073/pnas.96.26.15062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dong J, Joseph CA, Borotto NB, Gill VL, Maroney MJ, Vachet RW. Unique effect of Cu(II) in the metal-induced amyloid formation of beta-2-microglobulin. Biochemistry. 2014;53:1263–1274. doi: 10.1021/bi4016583. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Fan QR, Long EO, Wiley DC. Crystal structure of the human natural killer cell inhibitory receptor KIR2DL1-HLA-Cw4 complex. Nat Immunol. 2001;2:452–460. doi: 10.1038/87766. [DOI] [PubMed] [Google Scholar]
  11. Fan QR, Mosyak L, Winter CC, Wagtmann N, Long EO, Wiley DC. Structure of the inhibitory receptor for human natural killer cells resembles haematopoietic receptors. Nature. 1997;389:96–100. doi: 10.1038/38028. [DOI] [PubMed] [Google Scholar]
  12. Foley B, Felices M, Cichocki F, Cooley S, Verneris MR, Miller JS. The biology of NK cells and their receptors affects clinical outcomes after hematopoietic cell transplantation (HCT) Immunol Rev. 2014;258:45–63. doi: 10.1111/imr.12157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hou F, Sun L, Zheng H, Skaug B, Jiang QX, Chen ZJ. MAVS forms functional prion-like aggregates to activate and propagate antiviral innate immune response. Cell. 2011;146:448–461. doi: 10.1016/j.cell.2011.06.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Huse M, Eck MJ, Harrison SC. A Zn2+ ion links the cytoplasmic tail of CD4 and the N-terminal region of Lck. J Biol Chem. 1998;273:18729–18733. doi: 10.1074/jbc.273.30.18729. [DOI] [PubMed] [Google Scholar]
  15. Iannello A, Thompson TW, Ardolino M, Marcus A, Raulet DH. Immunosurveillance and immunotherapy of tumors by innate immune cells. Curr Opin Immunol. 2016;38:52–58. doi: 10.1016/j.coi.2015.11.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kagan JC, Magupalli VG, Wu H. SMOCs: supramolecular organizing centres that control innate immunity. Nat Rev Immunol. 2014;14:821–826. doi: 10.1038/nri3757. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kitamura H, Morikawa H, Kamon H, Iguchi M, Hojyo S, Fukada T, Yamashita S, Kaisho T, Akira S, Murakami M, Hirano T. Toll-like receptor-mediated regulation of zinc homeostasis influences dendritic cell function. Nat Immunol. 2006;7:971–977. doi: 10.1038/ni1373. [DOI] [PubMed] [Google Scholar]
  18. Kumar S, Sarkar P, Sim MJ, Rajagopalan S, Vogel SS, Long EO. A single amino acid change in inhibitory killer cell Ig-like receptor results in constitutive receptor self-association and phosphorylation. J Immunol. 2015;194:817–826. doi: 10.4049/jimmunol.1401830. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. LeVine H., 3rd Quantification of beta-sheet amyloid fibril structures with thioflavin T. Methods Enzymol. 1999;309:274–284. doi: 10.1016/s0076-6879(99)09020-5. [DOI] [PubMed] [Google Scholar]
  20. Li J, McQuade T, Siemer AB, Napetschnig J, Moriwaki K, Hsiao YS, Damko E, Moquin D, Walz T, McDermott A, et al. The RIP1/RIP3 necrosome forms a functional amyloid signaling complex required for programmed necrosis. Cell. 2012;150:339–350. doi: 10.1016/j.cell.2012.06.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Li YV. Zinc and insulin in pancreatic beta-cells. Endocrine. 2014;45:178–189. doi: 10.1007/s12020-013-0032-x. [DOI] [PubMed] [Google Scholar]
  22. Liu D, Peterson ME, Long EO. The adaptor protein crk controls activation and inhibition of natural killer cells. Immunity. 2012;36:600–611. doi: 10.1016/j.immuni.2012.03.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Long EO. Negative signaling by inhibitory receptors: the NK cell paradigm. Immunol Rev. 2008;224:70–84. doi: 10.1111/j.1600-065X.2008.00660.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Long EO, Kim HS, Liu D, Peterson ME, Rajagopalan S. Controlling natural killer cell responses: integration of signals for activation and inhibition. Annu Rev Immunol. 2013;31:227–258. doi: 10.1146/annurev-immunol-020711-075005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Moretta A, Bottino C, Vitale M, Pende D, Biassoni R, Mingari MC, Moretta L. Receptors for HLA class-I molecules in human natural killer cells. Annu Rev Immunol. 1996;14:619–648. doi: 10.1146/annurev.immunol.14.1.619. [DOI] [PubMed] [Google Scholar]
  26. Morvan MG, Lanier LL. NK cells and cancer: you can teach innate cells new tricks. Nat Rev Cancer. 2015;16:7–19. doi: 10.1038/nrc.2015.5. [DOI] [PubMed] [Google Scholar]
  27. Nguyen TA, Sarkar P, Veetil JV, Davis KA, Puhl HL, 3rd, Vogel SS. Covert Changes in CaMKII Holoenzyme Structure Identified for Activation and Subsequent Interactions. Biophys J. 2015;108:2158–2170. doi: 10.1016/j.bpj.2015.03.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Pageon SV, Aquino G, Lagrue K, Kohler K, Endres RG, Davis DM. Dynamics of natural killer cell receptor revealed by quantitative analysis of photoswitchable protein. Biophys J. 2013;105:1987–1996. doi: 10.1016/j.bpj.2013.09.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Pan E, Zhang XA, Huang Z, Krezel A, Zhao M, Tinberg CE, Lippard SJ, McNamara JO. Vesicular zinc promotes presynaptic and inhibits postsynaptic long-term potentiation of mossy fiber-CA3 synapse. Neuron. 2011;71:1116–1126. doi: 10.1016/j.neuron.2011.07.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Parham P, McQueen KL. Alloreactive killer cells: hindrance and help for haematopoietic transplants. Nat Rev Immunol. 2003;3:108–122. doi: 10.1038/nri999. [DOI] [PubMed] [Google Scholar]
  31. Rajagopalan S, Long EO. The direct binding of a p58 killer cell inhibitory receptor to human histocompatibility leukocyte antigen (HLA)-Cw4 exhibits peptide selectivity. J Exp Med. 1997;185:1523–1528. doi: 10.1084/jem.185.8.1523. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Rajagopalan S, Long EO. Zinc bound to the killer cell-inhibitory receptor modulates the negative signal in human NK cells. J Immunol. 1998;161:1299–1305. [PubMed] [Google Scholar]
  33. Rajagopalan S, Winter CC, Wagtmann N, Long EO. Cutting Edge: The Ig-related killer cell inhibitory receptor binds zinc and requires zinc for recognition of HLA-C on target cells. J Immunol. 1995;155:4143–4146. [PubMed] [Google Scholar]
  34. Sharma D, Masison DC. Single methyl group determines prion propagation and protein degradation activities of yeast heat shock protein (Hsp)-70 chaperones Ssa1p and Ssa2p. Proc Natl Acad Sci U S A. 2011;108:13665–13670. doi: 10.1073/pnas.1107421108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Siegel RM, Martin DA, Zheng L, Ng SY, Bertin J, Cohen J, Lenardo MJ. Death-effector filaments: novel cytoplasmic structures that recruit caspases and trigger apoptosis. J Cell Biol. 1998;141:1243–1253. doi: 10.1083/jcb.141.5.1243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Stebbins CC, Watzl C, Billadeau DD, Leibson PJ, Burshtyn DN, Long EO. Vav1 dephosphorylation by the tyrosine phosphatase SHP-1 as a mechanism for inhibition of cellular cytotoxicity. Mol Cell Biol. 2003;23:6291–6299. doi: 10.1128/MCB.23.17.6291-6299.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Steinhauser ML, Lechene CP. Quantitative imaging of subcellular metabolism with stable isotopes and multi-isotope imaging mass spectrometry. Semin Cell Dev Biol. 2013;24:661–667. doi: 10.1016/j.semcdb.2013.05.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Vivier E, Raulet DH, Moretta A, Caligiuri MA, Zitvogel L, Lanier LL, Yokoyama WM, Ugolini S. Innate or adaptive immunity? The example of natural killer cells. Science. 2011;331:44–49. doi: 10.1126/science.1198687. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Waggoner SN, Reighard SD, Gyurova IE, Cranert SA, Mahl SE, Karmele EP, McNally JP, Moran MT, Brooks TR, Yaqoob F, Rydyznski CE. Roles of natural killer cells in antiviral immunity. Curr Opin Virol. 2015;16:15–23. doi: 10.1016/j.coviro.2015.10.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Watzl C, Long EO. Natural killer cell inhibitory receptors block actin cytoskeleton-dependent recruitment of 2B4 (CD244) to lipid rafts. J Exp Med. 2003;197:77–85. doi: 10.1084/jem.20020427. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Wu H. Higher-order assemblies in a new paradigm of signal transduction. Cell. 2013;153:287–292. doi: 10.1016/j.cell.2013.03.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Yamasaki S, Sakata-Sogawa K, Hasegawa A, Suzuki T, Kabu K, Sato E, Kurosaki T, Yamashita S, Tokunaga M, Nishida K, Hirano T. Zinc is a novel intracellular second messenger. J Cell Biol. 2007;177:637–645. doi: 10.1083/jcb.200702081. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Yu M, Lee WW, Tomar D, Pryshchep S, Czesnikiewicz-Guzik M, Lamar DL, Li G, Singh K, Tian L, Weyand CM, Goronzy JJ. Regulation of T cell receptor signaling by activation-induced zinc influx. J Exp Med. 2011;208:775–785. doi: 10.1084/jem.20100031. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplement
NIHMS768317-supplement.pdf (4.3MB, pdf)

RESOURCES