Combination of CD8β Depletion and Interleukin-15 Superagonist N-803 Induces Virus Reactivation in Simian-Human Immunodeficiency Virus-Infected, Long-Term ART-Treated Rhesus Macaques

Abstract

The "shock and kill" strategy predicates that virus reactivation in latently infected cells is required to eliminate the human immunodeficiency virus (HIV) reservoir. In a recent study, we showed robust and persistent induction of plasma viremia in antiretroviral therapy (ART)-treated simian immunodeficiency virus-infected rhesus macaques (RMs) undergoing CD8α depletion and treated with the interleukin-15 (IL-15) superagonist N-803 (J. B. McBrien et al., Nature 578:154-159, 2020, https://doi.org/10.1038/s41586-020-1946-0). Of note, in that study we used an antibody targeting CD8α, thereby depleting NK cells, NKT cells, and γδ T cells, in addition to CD8⁺ T cells. In the current proof-of-concept study, we tested whether virus reactivation can be induced by administration of N-803 to simian-human chimeric immunodeficiency virus-infected, ART-treated RMs that are selectively depleted of CD8⁺ T cells via the CD8β-targeting antibody CD8b255R1. CD8β depletion was performed in five SHIV_SF162P3-infected RMs treated with ART for 12 months and with plasma viremia consistently below 3 copies/ml. All animals received four weekly doses of N-803 starting at the time of CD8b255R1 administration. The induction of detectable plasma viremia was observed in three out of five RMs, with the level of virus reactivation seemingly correlated with the frequency of CD8⁺ T cells following CD8β depletion as well as the level of virus reactivation observed when the same animals underwent CD8α depletion and N-803 administration after 24 weeks of ART. These data indicate that CD8β depletion and N-803 administration can induce virus reactivation in SHIV_SF162P3-infected RMs despite suboptimal depletion of CD8⁺ T cells and profound ART-induced suppression of virus replication, confirming a critical role for these cells in suppressing virus production and/or reactivation in vivo under ART.IMPORTANCE The "shock and kill" HIV cure strategy attempts to reverse and eliminate the latent viral infection that prevents eradication of the virus. Latency-reversing agents tested in clinical trials to date have failed to affect the HIV viral reservoir. IL-15 superagonist N-803, currently involved in a clinical trial for HIV cure, was recently shown by our laboratory to induce robust and persistent induction of plasma viremia during ART in three in vivo animal models of HIV infection. These results suggest a substantial role for CD8⁺ lymphocytes in suppressing the latency reversal effect of N-803 by promoting the maintenance of viral latency. In this study, we tested whether the use of a CD8β-targeting antibody, which would specifically deplete CD8⁺ T cells, would yield similar levels of virus reactivation. We observed the induction of plasma viremia, which correlated with the efficacy of the CD8 depletion strategy.

Keywords: AIDS; HIV; IL-15; N-803; SHIV; SIV; depletion; immunology; latency; reservoir.

FIG 1

Study design. Animals were infected intrarectally with high-dose SHIV_SF162P3, administered as a 1:50 dilution of a stock at 2,032 TCID₅₀/ml, 109 RNA copies/ml, and 182 ng/ml P27. All animals were put on a three-drug ART regimen at 12 weeks post-SHIV infection consisting of tenofovir (TDF at 5.1 mg/kg/day or PMPA at 20 mg/kg/day), emtricitabine (FTC at 40 mg/kg/day), and dolutegravir (DTG at 2.5 mg/kg/day). ART was administered daily by subcutaneous injection for the remainder of the study. After 24 weeks of ART, animals were administered one dose of MT807R1 (anti-CD8α) at 50 mg/kg subcutaneously. N-803 was administered subcutaneously in a cycle of 100 μg/kg once a week for four consecutive weeks starting at the time of CD8α depletion. Twenty-four weeks after administration of MT807R1 with N-803, animals were administered one dose of CD8b255R1 (anti-CD8β) at 50 mg/kg subcutaneously. N-803 was administered subcutaneously in a cycle of 100 μg/kg once a week for four consecutive weeks starting at the time of CD8β depletion. An elective necropsy was performed 7 weeks after administration of CD8b255R1 with N-803. Abbreviations: I.R., intrarectal; ART, antiretroviral therapy; N-803, IL-15 superagonist complex; NX, necropsy.

FIG 2

CD8β depletion with N-803 results in a modest virus reactivation correlated with CD8⁺ T cell frequencies but does not reduce the size of the SHIV reservoir. (A) Longitudinal plasma viral load starting at the time of infection, through ART initiation, CD8α depletion with N-803, CD8β depletion with N-803, and ending with animal necropsy. The limit of detection of the assay was 3 copies of SHIV RNA/ml of plasma. (B) Change in the level of cell-associated SHIV DNA in sorted, bulk, peripheral CD4⁺ T cells. Statistics were calculated using a Wilcoxon test. (C to E) Changes in the expression of HLA-DR (C), Ki67 (D), and PD-1 (E) on peripheral CD4⁺ T cells following intervention with CD8β depletion and N-803. Statistics were calculated using Friedman tests, and the level of significance for each comparison is indicated above the brackets (***, P < 0.001; **, P < 0.01; *, P < 0.05; nonsignificant [ns], P > 0.05). Each color designates a specific animal. Abbreviations: ART, antiretroviral therapy; NX, necropsy.

FIG 3

Reduction of CD8⁺ T cells following anti-CD8α or anti-CD8β. (A) Percentage of peripheral CD8⁺ cells from the CD3⁺ T cell population at the time of intervention (day 0) through day 49. (B) Longitudinal frequency of peripheral CD8⁺ T cells. (C to E) Percentage of CD8⁺ cells from the CD3⁺ T cell population at the start of intervention and after 7 days in the peripheral blood (C), lymph node (via a fine-needle aspirate) (D), and the rectum (via a rectal biopsy specimen) (E). (F) Fold change of CD8⁺ T cells 7 days after the start of intervention in the peripheral blood and rectum with CD8β depletion and CD8α depletion. This was calculated by dividing the percentage of CD8⁺ T cells at day 0 by that at day 7. Of note, fine needle aspirates of the lymph node were not collected following CD8α depletion. (G) The number of NK cells (CD3⁻ CD20⁻ CD14⁻ HLA-DR⁻ NKG2A⁺) present in peripheral blood longitudinally following the start of interventions. Intervention with anti-CD8β and N-803 is shown with solid lines or circles, and asterisks denoting statistical significance are black. Intervention with anti-CD8α is shown with dashed lines or squares, and asterisks denoting statistical significance are gray. Statistics for panels A, B, and G were calculated using Friedman tests and for panels C to F were calculated using two-sided Wilcoxon tests. The level of significance for each comparison is indicated above the brackets (****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *; P < 0.05; nonsignificant, P > 0.05). Each color represents a specific animal. Abbreviations: PBMC, peripheral blood mononuclear cells; LN, lymph node; RB, rectal biopsy specimen; ns, not significant.

FIG 4

Level of virus reactivation is correlated with the efficacy of CD8⁺ T cell depletion and previous depletions. (A and B) Correlation between the percentage of peripheral CD8⁺ T cells and plasma viral load (A) and peripheral CD8⁺ T cell count and viral load (B). Counts and viral loads from both depletion strategies were combined for the correlations. (C) Correlation between plasma viral loads following CD8α depletion and CD8β depletion. Correlations were calculated via Spearman’s rho. Each color designates a specific animal. Circles represent intervention using CD8β depletion with N-803, and squares represent intervention using CD8α depletion with N-803.