Development and optimization of human T-cell leukemia virus-specific antibody-dependent cell-mediated cytotoxicity (ADCC) assay directed to the envelope protein

Abstract

An estimated 10-20 million people worldwide are infected with the deltaretrovirus human T-cell leukemia virus type 1 (HTLV-1). Although most infected individuals remain asymptomatic, some progress to develop the fatal and debilitating disease adult T-cell leukemia/lymphoma (ATLL) or HTLV-associated myelopathy/tropical spastic paraparesis (HAM/TSP) or develop a plethora of other inflammatory disorders. In addition, HTLV-1 infection is associated with immunosuppression and a shorter lifespan. Although a protective role for neutralizing antibodies has been suggested, the role of non-neutralizing antibody-dependent cell-mediated cytotoxicity (ADCC) remains unclear, largely because an assay to measure this response has not been established. Here, we developed a high-throughput flow cytometry-based assay system to measure HTLV-1 envelope-specific ADCC. We used a natural killer cell-resistant T-lymphoblastoid cell line stably expressing the green fluorescent protein GFP to construct a target cell line expressing HTLV-1 envelope protein and using monoclonal antibodies and plasma samples from HTLV-infected or uninfected individuals, validating the specificity and sensitivity of the assay. We detected high ADCC activity in samples from HTLV-1-infected humans. In the plasma of experimentally infected macaques, ADCC activity was measured and a correlation between ADCC activity and HTLV-1 envelope antibody titers was observed. Further, we observed a significant increase in ADCC titer over time; as HTLV-1 infection persists, a higher ADCC response is generated, potentially influencing disease outcome. ADCC titer in HTLV-1-infected macaques also positively correlated with FLT3LG, IL-17F, CD4⁺ T cells, and lymphocytes but negatively correlated with monocyte frequency and classical monocyte frequency. In conclusion, these findings detail the generation of a cell line that enabled development of an HTLV-specific ADCC assay, which can be employed in large clinical studies as well as research involving humans or non-human primates.IMPORTANCEThis approach measures human T-cell leukemia virus (HTLV)-specific envelope antibody-dependent cell-mediated cytotoxicity responses, provides a critical tool to investigate the role of envelope-specific binding antibodies in the immune control of HTLV infection and pathogenesis, and may help guide the development of both therapeutic and preventative vaccine approaches.

Keywords: ADCC; CD16; Fc receptors; HTLV-1; adaptive immune response; natural killer cells.

Fig 1

Evaluation of HTLV-1 envelope-specific ADCC killing of established cell lines and characterization of new target cell line. (A and B) The HTLV-1 transformed, infectious, virus-producing cell lines MT2 and C91PL were labeled with cytoplasmic dye CFSE and membrane dye PKH26. Labeled cells were used as targets in a flow cytometry-based ADCC killing assay (see Materials and Methods). Graphed is the killing activity (percent) for MT2 and C91PL at 1 µg or 10 µg of the HTLV-1 monoclonal envelope antibody PRH7a. Background levels are indicated with a dotted line. Construction of CEM-NK^R-ENV1-EGFP cells. (C) Representative western blot analysis of the HTLV-1 envelope protein (anti-gp46, clone 65/6C2.2.34 Creative Biolabs, left or anti-HA, right) expression after transient transfection of the expression construct, pLv-ENV1, or control empty vector, labeled C, in 293T cells. (D) Stable CEM-NK^R-ENV1-EGFP clones (clonal line 1, ENV1a and clonal line 2, ENV1b) were selected, and expression of HTLV-1 envelope protein was assessed by western blot using anti-gp46 (left, clone 65/6C2.2.34 Creative Biolabs) or anti-HA (right). Whole cell extracts for the HTLV-1 positive cell lines C91PL and 729-D26 were used as controls for HTLV-1 envelope expression, and 729 and CEM-NK^R-EGFP (un) were used as negative controls. Anti-actin and anti-GFP antibodies were used for loading controls. (E) Representative flow cytometry plots for envelope expression (APC-Env) vs. cell size (SSC-A) are shown. A box around the envelope positive cells is shown on each plot.

Fig 2

Gating strategy and evaluation of ADCC killing in patient plasma. (A) The flow cytometry gating strategy for determining ADCC killing activity based on loss of GFP from the PKH-26 stained target cell in the presence or absence (background activity) of plasma. (B) ADCC activity against HTLV-1 envelope using monoclonal antibody (PRH7a) and CEM-NK^R-ENV1a-EGFP (green) or CEM-NK^R-ENV1b-EGFP (orange) cell lines as target cells. (C) A western blot analysis to determine seroreactivity to HTLV-1 proteins for plasma samples from HAM/TSP, ATLL, and uninfected (ND1) individuals. (D) The percentage of ADCC killing activity in the plasma for each HTLV-1-infected or uninfected (ND1 and ND2) individual was graphed. Activity was normalized to the monoclonal envelope antibody PRH7a, set to 100%. The dotted line indicates the background level. (E and F) Comparison of the percent (E) ADCC activity or (F) ADCC antibody titer in the plasma of HTLV-1-positive individuals and background (BG). Data shown in (E and F) were analyzed with two-tailed Mann-Whitney test. Horizontal and vertical bars denote mean and SD, respectively.

Fig 3

Envelope-specific ADCC activity in plasma from infected macaques. (A) ADCC killing activity from HTLV-1-infected macaques (green bars), HTLV-2-infected macaques (purple bars), or uninfected control (ZN19; white bar). The dotted line indicates the background level. (B and C) Comparison of (B) ADCC activity and (C) ADCC titer in HTLV-infected macaques from week 16 post-infection with background (BG). (D) Correlation between ADCC Ab titer and HTLV-1 p24 (Gag) Ab titer. (E) Correlation between ADCC activity and envelope Ab titer. Data shown in (B and C) were analyzed with two-tailed Mann-Whitney test. Horizontal and vertical bars denote mean and SD, respectively. Data shown in (D and E) were analyzed with two-tailed Pearson’s and two-tailed Spearman’s correlation test, respectively. Here, HTLV-1 and HTLV-2 samples are shown in green and brown, respectively.

Fig 4

Evaluation of ADCC activity and titer in plasma from HTLV-1A versus HTLV-1A/C_O1-L infected macaques. (A) Schematic diagram of the study design for infection of macaques with HTLV-1A (black, n = 5) or HTLV-1A/C_O1-L (red and blue, n = 4 each). The HTLV-1A triple depletion group and HTLV-1A/C_O1-L triple depletion group received M-T807R1 antibodies (gray arrows) for three consecutive days before viral challenge and clodrosome 1 day (red arrow) before viral challenge. The HTLV1-A/C_O1-L no depletion group respectively received IgG (purple arrows) and Liposome (green arrow) at the same time point. The black or blue arrow indicates intravenous injection of HTLV-1A or HTLV1-A/C_O1-L virus-producing lethally irradiated cells, respectively. (B) Comparison of ADCC killing activity in all animals over the course of the study. (C–E) Comparison of ADCC killing activity among triple depleted HTLV-1A (TD-WT, black), triple depleted HTLV1-A/C_O1-L (TD- A/C_O1-L, red), and non-depleted HTLV1-A/C_O1-L (ND- A/C_O1-L, blue) animals at weeks (C) 5, (D) 12, and (E) 21 post-infection. (F) Comparison of ADCC titer in all animals over the course of the study. (G–I) Comparison of ADCC titer among triple depleted HTLV-1A (TD-WT, black), triple depleted HTLV1-A/C_O1-L (TD- A/C_O1-L, red), and non-depleted HTLV1-A/C_O1-L (ND- A/C_O1-L, blue) animals at weeks (G) 5, (H) 12, and (I) 21 post-infection. Data shown in (B and F) were analyzed with two-tailed Wilcoxon signed-rank test and (C–E, G–I) were analyzed with two-tailed Mann-Whitney test. Horizontal and vertical bars denote mean and SD, respectively.

Fig 5

Correlation of ADCC activity at week 5 and week 12 post-infection with immune responses. (A and B) Negative correlations between ADCC titer and the frequency of (A) monocytes or (B) classical monocytes. (C and D) Positive correlations between ADCC titer and (C) FLT3LG or (D) IL17F cytokine plasma levels. (E and F) Correlation of total lymphocyte count with ADCC titer and ADCC killing activity, respectively. (G) Trend of positive correlation between ADCC killing activity (%) and total CD4⁺ cell count. Here, TD-WT (black), TD-ACO1-L (red), or ND-ACO1-L (blue). Data shown in (A–E) were analyzed with two-tailed Pearson’s correlation test and (F and G) were analyzed with two-tailed Spearman’s correlation test.

Fig 6

HTLV-specific antibody-dependent cellular cytotoxicity. Plasma from HTLV-infected macaques and human patients binds to HTLV antigen on the surface of CEM-NK^R-ENV1-EGFP target cells, which constitutively express GFP and plasma membrane labeled with PKH26. NK cells interact with the antibody-bound, HTLV-infected CEM-NK^R-ENV1-EGFP target cells through the FcγRIIIa receptor. This interaction triggers the release of perforin and granzyme from NK cells, forming pores in the target cells and leading to GFP release. The loss of GFP in PKH26-positive cells represents the fraction of cells killed through an antibody-dependent mechanism. This figure was created by F. Bhuyan with BioRender.com.