Elimination of SHIV Infected Cells by Combinations of Bispecific HIVxCD3 DART® Molecules

Abstract

Bispecific HIVxCD3 DART molecules that co-engage the viral envelope glycoprotein (Env) on HIV-1-infected cells and the CD3 receptor on CD3+ T cells are designed to mediate the cytolysis of HIV-1-infected, Env-expressing cells. Using a novel ex vivo system with cells from rhesus macaques (RMs) infected with a chimeric Simian-Human Immunodeficiency Virus (SHIV) CH505 and maintained on ART, we tested the ability of HIVxCD3 DART molecules to mediate elimination of in vitro-reactivated CD4+ T cells in the absence or presence of autologous CD8+ T cells. HIVxCD3 DART molecules with the anti-HIV-1 Env specificities of A32 or 7B2 (non-neutralizing antibodies) or PGT145 (broadly neutralizing antibody) were evaluated individually or combined. DART molecule-mediated antiviral activity increased significantly in the presence of autologous CD8+ T cells. In this ex vivo system, the PGT145 DART molecule was more active than the 7B2 DART molecule, which was more active than the A32 DART molecule. A triple combination of the DART molecules exceeded the activity of the individual PGT145 DART molecule. Modified quantitative virus outgrowth assays confirmed the ability of the DART molecules to redirect RM CD3+ T cells to eliminate SHIV-infected RM CD4+ T cells as demonstrated by the decreased propagation of in vitro infection by the infected cells pre-incubated with DART molecules in presence of effector CD8+ T cells. While mediating cytotoxic activity, DART molecules did not increase proinflammatory cytokine production. In summary, combination of HIVxCD3 DART molecules that have broadly-neutralizing and non-neutralizing anti-HIV-1 Env specificities can leverage the host immune system for treatment of HIV-1 infection but will require appropriate reactivation of the latent reservoir.

Keywords: HIV; bispecific DART molecules; broadly neutralizing antibodies; cytotoxic T cells; non-neutralizing antibodies; redirected cytotoxicity.

Figure 1

DART molecule binding to SHIV-infected A66 cells and their redirected killing in the presence of human CD8 T cells. The binding of HIVxRSV DART molecules with A32, 7B2 or PGT145 anti-HIV-1 Env specificities to SHIV.CH505.375H-infected A66 cells was evaluated to determine (A) frequency of SHIV-infected cells bound by DART molecules (%p27+/DART+) and (B) median fluorescent intensity (MFI). The 4420 DART molecule, which contains an anti-fluorescein specificity instead of an anti-HIV-1 Env specificity, was used as the negative control. (C) Titration curves represent DART-mediated killing of SHIV.CH505.375H-infected A66 cells as targets (T) by human CD8+ T cells as effectors (E) with E:T ratio of 33:1. The % Specific killing was determined 6 hours post by incubation of T+E+DART molecules measuring the reduction in the percentage of p27+ cells in presence of DART molecules compared to T+E alone.

Figure 2

Ex vivo killing assays with SHIV-infected RM CD4+ T cells and autologous RM CD8+ T cells. (A) Schematic of the in vivo study. Rhesus macaques (RMs) were infected intravenously (i.v.) with SHIV.CH505.375H at a dose of 10 ng of SIV Gag p27 antigen (41). SHIV.CH505.375H is a chimeric simian/human immunodeficiency virus with SIVmac backbone and HIV-1 transmitted/founder clade C envelope CH505. The RMs were treated with an ART regimen consisting of tenofovir (TDF: 5.1 mg/kg), emtricitabine (FTC: 40 mg/kg), and dolutegravir (DTG: 2.5 mg/kg) administered subcutaneously daily from week 16 post-infection (p.i.) to the end of the study. PBMCs were collected from the SHIV-infected RMs at week 2 post-infection (prior to ART) to isolate SHIV-infected CD4+ T cells (targets) and at week 36 post-infection (while on ART) to isolate CD8+ T cells (effectors). (B) Design of the ex vivo killing assays. Primary CD4+ T cells from SHIV.CH505.375H-infected RMs were isolated and activated in vitro for 24 hours with antibodies specific for nonhuman primate CD2/CD3/CD28. Activated SHIV-infected RM CD4+ T cells were cultured for 48 hours alone or with autologous RM CD8+ T cells at an E:T ratio of 0.03:1 in the absence or presence of DART molecules. The % killing of the SHIV-infected RM CD4+ T cells was determined by measuring the reduction in supernatant SIV Gag p27 levels compared to CD4+ T cells alone.

Figure 3

Virologic parameters. (A) Plasma viral load (SHIV RNA copies/mL) in individual animals (indicated by their ID designations) following 2 weeks post infection with SHIV.CH505.375H. (B) Cell-associated viral DNA (caDNA) and (C) cell-associated viral RNA (caRNA) prior to ART initiation at week 16 post-infection. (D) SIV Gag p27 levels in supernatants of primary RM CD4+ T cells collected at week 2 post-infection, activated for 24 hours in vitro with anti-CD2, CD3, and CD28 antibodies and cultures for 3 days in fresh media. (A–D) In the box plots, horizontal lines within boxes denote median values, ends of boxes denote 25^th and 75^th percentiles, and lines outside of boxes denote minimum and maximum values. (E) Spearman correlation analysis of p27 levels in supernatants following in vitro activation of RM CD4+ T cells, peak plasma viral load, cell-associated viral RNA, and cell-associated viral DNA. Values are R² values.

Figure 4

HIVxCD3 DART molecule-mediated killing of SHIV-infected RM CD4+ T cells. In vitro activated primary CD4+ T cells from SHIV.CH505.375H-infected RMs were cultured without or with DART molecules in the absence (A) or presence (B) of autologous CD8+ effector cells at an E:T ratio of 0.03:1. The A32, 7B2 and PGT145 DART molecules at 1 µg/mL were added individually or as a triple combination (Combo) at 1 µg/mL each. The reduction in p27 levels (A) and DART-mediated killing (B) was analyzed by p27 ELISA in culture supernatants 48 hours post treatment. The circles represent cultures with CD4 cells, squares represent cultures with CD4 and CD8 cells. Each colored symbol represents an individual animal. The horizontal black bars represent median values. Statistical analysis of p27 levels (C) or % killing (D) between indicated groups was performed using Wilcoxon Test at the significance level of 0.05. NS, not significant.

Figure 5

Effect of DART molecules on virus propagation. Cultures of reactivated SHIV-infected RM CD4+ T cells alone (CD4 cells) or mixed with autologous RM CD8+ T cells and incubated in the absence (No DART) or presence of DART molecules for 48 hours. DART molecules were then washed away and the cultures were mixed with A66 feeder cells to allow virus propagation. On day 9 cultures supernatants were analyzed for SIV Gag p27 levels by ELISA (A). The horizontal lines represent the median values. (B) Statistical comparison between the indicated groups was performed using the Wilcoxon test at a significance level of 0.05. Combo is a combination of A32, 7B2 and PGT145 DART molecules.