HIV-gp140-Specific Antibodies Generated From Indian Long-Term Non-Progressors Mediate Potent ADCC Activity and Effectively Lyse Reactivated HIV Reservoir

Abstract

Strategies to reduce the human immunodeficiency virus (HIV) reservoir are urgently required. The antibody-dependent cellular cytotoxicity (ADCC)-mediating anti-HIV antibodies have shown an association with HIV control. We assessed if such antibodies can be generated in vitro and whether the generated antibodies can facilitate the reduction of reactivated HIV reservoir. We isolated HIV-1-gp140-specific memory B cells from HIV-1-infected long-term non-progressors (LTNPs) with or without plasma ADCC and cultured them to generate anti-HIV antibodies. The ability of the generated antibodies to mediate ADCC and facilitate NK cell-mediated lysis of reactivated HIV reservoir was assessed by the rapid fluorometric antibody-dependent cellular cytotoxicity assay and a flow-based novel latency reduction assay, respectively. All LTNPs showed the presence of gp140-specific memory B cells [median: 0.79% (0.54%-1.225%)], which were successfully differentiated into plasma cells [median 72.0% (68.7-82.2%)] in an in-vitro culture and secreted antibodies [median OD: 0.253 (0.205-0.274)]. The HIV-gp140-specific antibodies were generated from 11/13 LTNPs irrespective of their plasma ADCC status. The generated antibodies from LTNPs with plasma ADCC showed higher ADCC potency (median: 37.6%, IQR: 32.95%-51%) and higher reduction in reactivated HIV reservoir (median: 62.5%, IQR: 58.71%-64.92%) as compared with the antibodies generated from LTNPs without plasma ADCC (ADCC: median: 8.85%, IQR: 8%-9.7%; and % p24 reduction median: 13.84, IQR: 9.863%-17.81%). The potency of these antibodies to reduce latent reservoir was two-fold higher than the respective plasma ADCC. The study showed that the potent ADCC-mediating antibodies could be generated from memory B cells of the LTNPs with plasma ADCC activity. These antibodies also showed potent ability to facilitate NK cell-mediated lysis of reactivated HIV reservoirs. It also indicated that memory B cells from individuals with plasma ADCC activity should be preferentially used for such antibody generation. The important role of these antibodies in the reduction of latent reservoirs needs to be further evaluated as a useful strategy to obtain a functional cure for HIV infection.

Keywords: HIV-gp140 antibodies; HIV-specific memory B cells; antibody-dependent cellular cytotoxicity (ADCC); long-term non-progressors; reactivated latent reservoirs.

Figure 1

Summary of the experimental workflow of the study. (A) Whole blood samples were collected from long-term non-progressors (LTNPs) with and without plasma antibody-dependent cellular cytotoxicity (ADCC) activity (R-ADCC, n = 10 and NR-ADCC, n = 3, respectively) and 3 healthy individuals (HCs) and processed for the separation and cryopreservation of peripheral blood mononuclear cells (PBMCs). The PBMCs were then revived and memory B cells from three HCs and HIV-specific memory B cells from R-ADCC (n = 10) and NR-ADCC LTNPs (n = 3) were FACS sorted. (B) Antibody generation: the FACS-sorted memory B cells from all the samples were cultured for 10 days. On day 10, supernatants were assessed for the presence of antibodies and were then column purified. The cells were assessed for the differentiation into plasma cells by flow cytometry and real-time PCR analysis at day 10. (C) Antibody characterization: the generated antibodies were then assessed for the presence of gp140 antibodies using ELISA, their ability to mediate ADCC using rapid fluorometric antibody-dependent cellular cytotoxicity (RF-ADCC) assay, and their ability to facilitate NK cell-mediated lysis of Env-reactivated latent reservoirs using a flow-based latency reduction assay as reported earlier (16).

Figure 2

Generation of HIV-specific antibodies from HIV-specific memory B cells. (A) The dot plots show the gating strategy used to identify gp140+ memory B cells. The CD3− cells were gated from the lymphocytes which were then drilled down to gate CD3−CD19+CD27+ memory B cells. (B) To assess the binding specificity of the gp140 probe to a specific BCR, the PBMCs were co-stained with an HIV-gp140 protein probe and one of the three non-HIV antigen probes. The antigen-specific memory B cells were gated from CD3−CD19+CD27+ memory B cells. Subpanel (B1) represents the fluorescence minus one (FMO) control and subpanels (B2-B4) represent the dot plots showing the cells stained by biotinylated HIV-1 gp140 probe-conjugated with streptavidin PE-Cy7 (X-axis in all) and the streptavidin BV510 tagged biotinylated antigens, viz., H1N1 antigen, Rubeola antigen, and tetanus antigen, respectively (Y-axis). (C) From the CD3−CD19+CD27+ memory B cells, the gp140+ memory B cells were identified as CD3−CD19+CD27+gp140+ memory B cells. (D) The dot plots show the gating strategy used for plasma cells. 1) The live cells were gated from lymphocytes which were further drilled down to gate CD20−CD38+ plasmablast cells. 2) Further from these cells, plasma cells were gated as CD20−CD38+CD138+ cells (right panel) using FMO control. (E) The bar diagram depicts the frequencies of gp140-specific memory B cells (Y-axis) in R-ADCC LTNPs (n = 10), NR-ADCC LTNPs (n = 3), and HCs (n = 3) (X-axis) at the initiation of culture. (F) The cells at day 0 and day 10 of culture were analyzed for the expression of TFs of BCL-6, PAX-5, XBP-1, and PRDM-1 genes using real-time PCR. The expression levels of mRNA were normalized with β-actin and fold change (ΔΔCT) (plotted on the Y-axis) was calculated with the expression of the respective genes at day 0 as baseline. The data represent the mean expression level from eight samples. The error bar indicates the standard error of the mean. (G) The scattered plot shows the concentrations of IgGs (μg/ml) (Y-axis) in the supernatants collected on the 4th and 10th day of culture (X-axis). (H) The concentration of gp140-binding antibodies (µg/ml) (Y-axis) generated from the gp140+ memory B cells of R-ADCC LTNPs (R1 to R10), NR-ADCC LTNPs (NR1 to NR3), and HIV-negative individuals (X-axis). The * represents P ≤ 0.05 and ** represents P ≤ 0.01.

Figure 3

Assessment of ADCC activity of the generated antibodies. (A) The dot plots represent the gating strategy used for the RF-ADCC assay. The CFSE^neg cells (2) were gated from FSC/SSC scatters (1) which were drilled down to gate CD3− cells (3). The CD14+ monocytes (4) were then gated from CD3− cells and further analyzed for PKH expression. The histogram shows the comparison of PKH26 uptake by CD14⁺ monocytes in the presence of purified antibodies generated from gp140-specific memory B cells from R-ADCC LTNPs (R2: green line, R4: blue line, and R5: orange line), HIV-seronegative (dotted line), and in PBMC+CEM control (red line) (5). (B) The box and whiskers plots (left panel) depict the ADCC activity (expressed in terms of % PKH uptake by monocytes) against HIV-1 Env C protein (left panel) (Y-axis) in R-ADCC (n = 10) and NR-ADCC (n = 3) LTNPs and HIV-negative individuals (n = 3) (X-axis). Boxes show median (IQR). The line diagram (right panel) shows the paired data of ADCC activity (Y-axis) of purified antibodies and their respective unbound fractions from R-ADCC LTNPs (X-axis) against HIV-1 Env C protein (right panel). (C) The scatter dot plot (lines showing the median with IQR) shows the ADCC activity (Y-axis) against Env C and Env B protein (X-axis) of purified antibodies generated from gp140-specific memory cells of R-ADCC LTNPs (n = 10). The Wilcoxon matched-pairs t-test was used for the data analysis and p ≥0.05 was considered to be significant for all the analyses. The ** represents P ≤ 0.01.

Figure 4

Gating strategy used to assess killing of reactivated HIV-1 latent reservoirs by the generated ADCC antibodies. To assess the ability of ADCC antibodies to lyse reactivated HIV latent reservoirs, the FACS-sorted resting memory CD4+ T cells (target) were reactivated with Env protein and co-cultured with CD4−CD8− cell (effector) in the presence of generated antibodies. The reduction in the frequencies of p24-expressing resting memory CD4+ T cells was used as the final readout of the assay. The dot plots depict the representative graphs for (A) the frequencies of p24-expressing resting memory CD4+ T cells. The live cells were gated from lymphocytes by using the live-dead dye. From the live cells, CD3+CD4+ cells were gated which were further drilled down to gate CD4+CD45RO+ resting memory T cells. These cells were then assessed for the expression of p24 in different culture conditions, viz., uninfected, unstimulated, Env-stimulated, and after the addition of antibodies from R-ADCC and NR-ADCC LTNPs. The uninfected cells were kept as a control to set the gate. (B) The dot plots show the gating strategy used to assess the simultaneous activation of CD3−CD56^dim NK cells (measured in terms of secretion of cytokine IFNγ and degranulation marker CD107a) in different culture conditions, viz., uninfected, unstimulated, Env-stimulated, and after the addition of IgGs from R-ADCC and NR-ADCC LTNPs.

Figure 5

Assessment of killing of reactivated HIV-1 latent reservoirs by the generated ADCC antibodies. (A) The line diagram shows the % p24-expressing CD4+CD45RO+ resting memory T cells (Y-axis) in unstimulated cells, after Env C stimulation and after the addition of antibodies generated from R-ADCC LTNPs (black dots) and NR-ADCC LTNPs (red dots) to the Env-stimulated cells (X-axis). (B) The scatter plot shows % p24 reduction (Y-axis) in five R-ADCC LTNPs (R1, R3, R4, R6, and R7) and 2 NR-ADCC LTNPs (NR1 and NR2) (X-axis). (C) The box and whiskers plots % NK cell activation (Y-axis) in unstimulated cells, after Env C stimulation and after the addition of antibodies generated from R-ADCC LTNPs and NR-ADCC LTNPs to the Env-stimulated cells (X-axis). The Wilcoxon matched-pairs t-test was used to analyze the data and p ≥0.05 was considered to be significant. The ** represents P ≤ 0.01.

Figure 6

Comparison of potency to reduce latent reservoirs of generated antibodies and paired plasma antibodies. The bar diagram represents the comparative analysis of reduction in percent p24-expressing resting memory CD4+ T cells in the presence of antibodies generated from R-ADCC (n = 5) LTNPs and purified plasma antibodies of the same LTNPs (n = 5) (paired t-test, p = 0.0016). The ** represents P ≤ 0.01.