The role of natural killer (NK) cells and NK cell receptor polymorphisms in the assessment of HIV-1 neutralization

Abstract

The importance of innate immune cells in HIV-1 pathogenesis and protection has been highlighted by the role of natural killer (NK) cells in the containment of viral replication. Use of peripheral blood mononuclear cells (PBMC) in immunologic studies provides both HIV-1 target cells (ie. CD4+ T cells), as well as anti-HIV-1 effector cells, such as NK cells. In this study, NK and other immune cell populations were analyzed in HIV-negative donor PBMC for an impact on the anti-HIV activity of polyclonal and monoclonal antibodies. NK cell percentages were significantly higher in donor PBMC that supported lower levels of viral replication. While the percentage of NK cells was not directly associated with neutralization titers, NK cell-depletion significantly diminished the antiviral antibody activity by up to three logs, and polymorphisms in NK killer immunoglobulin receptor (KIR) and FcγRIIIa alleles appear to be associated with this affect. These findings demonstrate that NK cells and NK cell receptor polymorphisms may influence assessment of traditional HIV-1 neutralization in a platform where antibody is continuously present. This format appears to simultaneously assess conventional entry inhibition (neutralization) and non-neutralizing antibody-dependent HIV inhibition, which may provide the opportunity to delineate the dominant antibody function(s) in polyclonal vaccine responses.

Figure 1. Donors with more NK cells have decreased viral growth.

Donors were stratified by viral growth (A, n=25) and neutralization (B, n=19) then separated into quartiles. The percentages of NK cells for each donor were determined from an average of 3 experiments using flow cytometry and plotted based on quartile. The bars represent median. The Kruskal-Wallis test was used to determine statistical significance.

Figure 2. NK cells participate in viral inhibition.

PBMC from 4 donors were depleted of NK cells (white bars), and then compared with matched bulk PBMC as targets (black bars) in neutralization assays. The IC₈₀s (mAbs, panel A) or ID₈₀s (polyclonal plasma, panel B) are indicated. Each bar represents the average of 2 experiments and the error bars show the standard error of the mean.

Figure 3. Depletion of NK cells results in a significant loss of neutralization.

The ID₈₀s for Bulk PMBC and matched NK-depleted PBMC for 18 donors are displayed (A). Also displayed is the fold-loss of neutralization after NK cell depletion comparing donors who are 3DS1+ and 3DS1− (B) and those who are FcγRIIIa158V+ vs. 158V− (C). The black lines represent the median values. Each neutralization titer is an average of 2 experiments. The Wilcoxon rank test was used to determine statistical significance.

Figure 4. Donor PBMC ranked by neutralization.

PBMC from 19 donors were used as target cells for neutralization using 6 virus stocks and 7 neutralization reagents. The ranking and separation of donors into the upper two quartiles (“High neutralization") and the lower two quartiles (“Low neutralization") is indicated, along with the genotypes for KIR3D, HLA-B_80, and FcγRIIIa 158. Red text indicates 3DS1+, Valine (V) at FcγRIIIa position 158, or licensing through HLA-B with Bw4-80I or Bw4-80T.

Figure 5. Target PBMC from KIR3DS1+ and FcgRIIIa 158V+ donors show higher levels of neutralization than KIR3DS1− and FcgRIIIa 158V−.

The IC₅₀s for the mAbs and sCD4, and the ID₅₀s for the polyclonal antibodies, are displayed for NL-LucR.T2A-BaL.ecto (A and C) and NL-LucR.T2A-SF162.ecto (B and D). Donors were separated by the KIR3DS genotype (A and B) and the FcγRIIIa 158V genotype (C and D). The inter-quartile ranges for the KIR3DS1− and FcγRIIIa 158V− donors are displayed in blue bars while the inter-quartile range for the KIR3DS1+ and FcγRIIIa 158V+ donors are displayed in red bars. The black lines within each group represent the median neutralization values, while the grey symbols represent individual neutralization titers. Each neutralization titer is an average of 2 experiments.

Figure 6. NK cell depletion results in loss of neutralization against a CRF01_AE virus when using a ICp24 assay format.

Neutralization was calculated employing a subtype CRF01_AE IMC (virus CM235) without a reporter gene inserted and using a flow cytometric intracellular p24 endpoint (panels A–D). The ID₅₀s for 2 polyclonal antibody pools (A, C) or IC₅₀s for 4E10 (B, D) are displayed for PBMC from 2 donors that were used either in bulk (black bars) or after NK cell depletion (white bars).

Figure 7. Higher neutralization is observed with antibody maintained in the assay.

Using the 2 lucifease reporter IMCs (NL-LucR.T2A-SF162.ecto and NL-LucR.T2A-BaL.ecto), the ID₅₀s for a serum pool (USHIV+) and 2 individual plasmas are displayed for assay formats where the antibodies were maintained throughout the culture period (Ab maintained, black bars) or removed after infection (Ab removed, white bars). Each neutralization titer is an average of 2 experiments.

Figure 8. Model for the engagement of NK cells to mediate HIV inhibition.

HIV infection and growth may be suppressed in a variety of ways. 1) “Traditional neutralization" where the antibodies bind to virions and inhibit virion attachment and/or fusion with target cells. 2) ADCC, in which antibodies bind to viral proteins on the surface of infected cells, enabling NK cells to engage and kill the target cells. 3) ADCVI, which includes not only ADCC, but also the secretion of cytokines and/or chemokines that inhibit infection. 4) Direct killing of infected cells triggered by downregulation of HLA_A and HLA_B, loss of inhibitory signaling, and subsequent activation of NK killing. Polymorphisms in the FcγRIIIa may affect functions shown in 2 and 3; polymorphisms in KIR3D may affect NK cell function diagrammed in 4.