A novel ligand on astrocytes interacts with natural cytotoxicity receptor NKp44 regulating immune response mediated by NK cells

Abstract

NK cells play important role in immunity against pathogens and cancer. NK cell functions are regulated by inhibitory and activating receptors binding corresponding ligands on the surface of target cells. NK cells were shown to be recruited to the CNS following several pathological conditions. NK cells could impact CNS physiology by killing glial cells and by secreting IFN-γ. Astrocytes are intimately involved in immunological and inflammatory events occurring in the CNS and reactive astrogliosis is a key feature in HIV-associated neurocognitive disorders. There is little data on NK-astrocyte interactions and ligands expressed on astrocytes that could impact NK cell function. Natural cytotoxicity receptors (NCRs) play a critical role in the cytolytic function of NK cells. Among the NCRs, NKp44 is unique in expression and signal transduction. NKp44 is expressed only upon activation of NK cells and it can mediate both activating and inhibitory signals to NK cells. Here, we have studied the expression and function of natural cytotoxicity receptor NKp44 upon NK-astrocytes interactions in the presence or absence of an HIV peptide (HIV-3S peptide) shown to induce NK cell killing of CD4+ T cells during HIV-infection. Using a fusion protein consisting of the extracellular domain of NKp44 fused to Fc portion of human IgG, we determined the expression of a novel ligand for NKp44 (NKp44L) on astrocytes. Incubation of astrocytes with HIV-3S peptide downregulated NKp44L expression on astrocytes implicating protection from NK mediated killing. Thus, our study showed that NKp44 have a protective effect on astrocytes from NK cell mediated killing during HIV infection and impact astrocyte role in HAND.

Fig 1. Schematic representation of the mammalian expression vector construct and the NKp44-Fc fusion protein.

(A) Fusion protein is bivalent, containing two extracellular domains in the arm regions and one Fc region. (B) The extracellular domain of NKp44 was PCR amplified and subcloned in front of the CH2, CH3 domain of human IgG1 in frame with the CD5 leader sequence and CH2 as described in materials and methods.

Fig 2. Astrocytes express a ligand for NKp44 and increasing concentrations of HIV 3S peptide decreases its expression.

(A) The expression of a ligand for NKp44 (NKp44L) on astrocytes was established using 70 μg of NKp44-Ig fusion protein detected by anti-IgG-Fc-PE (filled histogram). Astrocytes incubated with 2.5 μg anti-IgG-Fc-PE were used as negative controls (open histogram). Cells were tested for the binding of NKp44-Ig fusion protein with no stimulation, 1 μg/ml peptide stimulation, and 10 μg/ml stimulation. Mean Fluorescent Intensity (MFI) values, the differences between the mean value for the control group and the experimental groups were quantified and MFI is indicated in the top right corner of all plots. Fig A is representative of 1 sample. (B) The combined MFIs of tests done in duplicate on 5 different astrocyte donors. Data are represented as means ± SEM, **** p < 0.0001, ANOVA.

Fig 3. Blocking NKp44 interaction inhibits NK92-MI and primary NK cell mediated killing of astrocytes.

NK92-MI and primary NK cell mediated lysis of astrocytes was determined using a standard ⁵¹Cr release assay. Astrocytes and NK92-MI cells (A & B) / primary NK cells (C & D) were first blocked with Human IgG Fc fragment to prevent reverse binding of fusion protein and antibody dependent cellular cytotoxicity. Astrocytes cells were labeled with ⁵¹Cr and incubated with 1 μg/μl of NKp44-Ig fusion protein (A, C). Astrocytes were then incubated with NK92-MI and primary NK cells at varying effector to target cell ratios for 4 hours at 37°C. Percent specific lysis of astrocytes was compared to astrocytes incubated with 0.5 mg/ml mIgG1 isotype antibody or no antibody, which served as a positive control (No Blocking) of cell lysis under unblocked conditions. Alternatively, NK92-MI and primary NK cells were incubated with 0.5 mg/ml anti-NKp44 or mIgG1 isotype control antibody prior to incubation with astrocytes incubated with no antibody (B, D). Figure is representative of 3 independent experiments performed in triplicate. Data is displayed as means ± SEM. ** p < 0.01, *** p < 0.001, **** p < 0.0001, ANOVA.

Fig 4. HIV-3S peptide stimulation protects astrocytes and the blocking of NKp44 further protects HIV-3S stimulated astrocytes from NK92-MI and primary NK cell mediated killing.

Astrocytes were stimulated with 10 μg/ml HIV 3S peptide for 4 hours prior being labeled with ⁵¹Cr. NK92-MI and primary NK cell mediated lysis of astrocytes was determined using a standard ⁵¹Cr release assay. Astrocytes and NK92-MI cells (A & B) / primary NK cells (C & D) were first blocked with Human IgG Fc fragment to prevent reverse binding of fusion protein and antibody dependent cellular cytotoxicity. Astrocytes cells were labeled with ⁵¹Cr and incubated with 1 μg/μl of NKp44-Ig fusion protein (A, C). Astrocytes were then incubated with NK92-MI and primary NK cells at varying effector to target cell ratios for 4 hours at 37°C. Percent specific lysis of astrocytes was compared to astrocytes incubated with 0.5 mg/ml mIgG1 isotype antibody or no antibody, which served as a positive control (No Blocking) of cell lysis under unblocked conditions. Alternatively, NK92-MI and primary NK cells were incubated with 0.5 mg/ml anti-NKp44 or mIgG1 isotype control antibody prior to incubation with astrocytes incubated with no antibody (B, D). Figure is representative of 3 independent experiments performed in triplicate. Data is displayed as means ± SEM. ** p < 0.01, *** p < 0.001, **** p < 0.0001, ANOVA.

Fig 5. Blocking NKp44 interaction decreases the release of IFN-γ.

IFN-γ levels were determined from astrocyte-NK cell co-culture supernatants by an IFN-γ ELISA. NK92-MI cells and astrocytes were first Fc blocked. NK92-MI cells were then treated with 2.5 μg anti-NKp44 and then incubated with astrocytes in the absence (A) and presence of HIV-3S peptide (B). NK92-MI cells with no antibody treatment or with 2.5 μg mIgG1 were used as controls. Astrocytes with and without HIV-3S peptide stimulation was treated with 1 μg anti-NKp44, mIgG1 isotype control or received no antibody treatment. Figure is representative of 2 independent experiments performed in triplicate. Bars ± SEM, ** p < 0.01, ANOVA.