Glycoprotein L-deleted single-cycle rhesus cytomegalovirus vectors elicit MHC-E-restricted CD8+ T cells that protect against SIV

Abstract

Strain 68-1 rhesus CMV (RhCMV) vectors induce immune responses that mediate early, complete replication arrest of SIV infection in ∼60% of vaccinated rhesus macaques (RMs). This unique efficacy depends on the ability of these vectors to elicit effector memory (EM)-biased CD8+ T cells recognizing SIV peptides presented by MHC-E, rather than MHC-Ia. These efficacious responses still occurred when spread of the 68-1 vector was impaired by deletion of the viral anti-host intrinsic immunity factor phosphoprotein 71 (pp71), but efficacy was lost with a more stringent attenuation strategy based on destabilization of Rh108, the ortholog of the essential human CMV (HCMV) transcription factor UL79 that is required for late viral gene expression. Although unable to produce infectious progeny (ie single-cycle infection), Rh108-deficient vectors elicited durable, high frequency, EM-biased, SIV-specific CD8+ T-cell responses in RMs, but these responses were MHC-Ia-restricted and therefore non-efficacious. Here, we tested a different single-cycle attenuation strategy based on deletion (Δ) of the glycoprotein L (gL) that is essential for viral entry but allows for late gene expression and viral assembly. ΔgL 68-1 RhCMV/SIV vectors, grown on gL-complementing fibroblasts, were robustly immunogenic at doses above 105 PFU, generating high frequency, EM-biased, SIV-specific CD8+ T-cell responses that were also unconventionally restricted, including the MHC-E restriction associated with efficacy. Indeed, these single-cycle vectors manifested replication arrest efficacy in 70% of vaccinated RMs, further linking MHC-E restriction with efficacy, and demonstrating that 68-1 RhCMV/SIV efficacy does not require vector dissemination within the host.

Keywords: CD8+ T cells; CMV vectors; HIV vaccine; MHC-E.

Figure 1.

Design and validation of ΔgL RhCMV/SIV vaccine vectors. (A) Schema of the genetic modifications made in the 68–1 RhCMV backbone to produce the ΔgL RhCMV/SIV vector. (B) Immunofluorescence analysis of unmodified telomerized rhesus fibroblasts (TRFs) and RhCMV gL–expressing TRFs (TRF-gL) using a polyclonal antibody to RhCMV gL. The TRF-gL cells were used to generate and produce ΔgL RhCMV/SIV vectors. (C) Multistep growth curve (multiplicity of infection = 0.01) comparing the in vitro replication and spread of gL-intact 68–1 RhCMV (wild type: WT) with ΔgL RhCMV on unmodified TRF or TRF-gL. All virus titers in the supernatant were determined on TRF-gL. (D) In vivo dose response analysis of ΔgL RhCMV/SIVgag (1 RM for each dose) using the induction of SIV Gag–specific CD4⁺ and CD8⁺ T-cell responses to indicate take of vector infection. Responses were measured by ICS analysis for IFN-γ and TNF-α using a pool of 15-mer peptides overlapping by 11 amino acids corresponding to the SIVmac239 Gag protein sequence to stimulate T cells in PBMC. The frequency of IFN-γ and/or TNF-α positive memory T cells is shown for each of the indicated post-inoculation time points.

Figure 2.

Magnitude and durability of SIV-specific T-cell responses elicited by ΔgL vs. gL-intact 68–1 RhCMV/SIV vaccine vectors. (A) Schema of the animal protocol used to study the immunogenicity and efficacy of ΔgL (group 1) vs. gL-intact (group 2) 68–1 RhCMV/SIV vaccines. (B) Longitudinal analysis of the overall SIV-specific CD4⁺ and CD8⁺ T-cell responses in peripheral blood. Responses were determined by ICS analysis (IFN-γ and TNF-α) using whole open reading frame (ORF) mixes of overlapping 15-mer peptides (SIVmac239 Gag; Rev/Nef/Tat; 5’-Pol) to stimulate PBMCs. The frequencies of IFN-γ and/or TNF-α positive memory T-cell responses to each ORF peptide mix were summed to get the overall responses shown in the figure. The 2-sample Wilcoxon rank-sum (WRS) test was used to determine the significance of differences between ΔgL (group 1) and gL-intact (group 2) vaccination for the area under the curve of SIV-specific CD4⁺ T-cell responses for the time periods shown in the figure. (C) Plateau phase analysis of the overall SIV and individual SIV ORF-specific CD4⁺ and CD8⁺ T-cell responses in peripheral blood during pre-challenge plateau phase for group 1 (weeks 39 to 57) and group 2 (weeks 41 to 80) RMs, with each point representing the mean response across all time points in the designated time intervals. The 2-sample WRS test was used to determine the significance of differences between ΔgL (group 1) and gL-intact (group 2) vaccination. For the total responses unadjusted P-values were nonsignificant; for the separate antigens Holm-adjusted P-values were nonsignificant.

Figure 3.

Induction of unconventionally restricted CD8⁺ T-cell responses by ΔgL RhCMV/SIV vaccine vectors. (A) Longitudinal analysis of ΔgL RhCMV/SIV vaccine–induced CD8⁺ T-cell responses to individual MHC-E–restricted (green; left panel) and MHC-II–restricted (blue; right panel) 15-mer supertopes. Responses were determined by ICS as described in Fig. 2. (B) Comparative analysis of the ICS-determined magnitude of individual 15-mer supertope–specific responses during plateau phase (weeks 39 to 42 post–initial vaccination) for RMs in group 1 (ΔgL 68–1 RhCMV/SIV vaccinated) vs. group 2 (gL-intact 68–1 RhCMV/SIV vaccinated). The 2-sample WRS test (adjusted for multiple comparisons) was used to determine the significance of differences between group 1 and group 2. Holm-adjusted P-values were nonsignificant. (C–E) Plateau phase blood CD8⁺ T cells from 5 representative RMs inoculated with ΔgL 68–1 RhCMV/SIV were assessed by ICS (TNF-α and/or IFN-γ) for CD8⁺ T-cell responses to each consecutive (11 amino acids overlapping) 15-mer peptide shown, including 15-mers 1–25 for SIV Rev (C), 15-mers 30–70 for SIV Pol (D), and all 125 15-mers for SIV Gag (E), with any responses >0.05% (after background subtraction) indicated by a box. Boxes are colored to reflect MHC restriction based on inhibition of the response with MHC-E–blocking peptide, MHC-II–blocking mAb, and/or pan-MHC-I–blocking mAb.^– For each RM, the minimal number of epitopes of each class was calculated for each ORF or ORF segment, as previously described.^–

Figure 4.

Phenotypic and functional analysis of SIV Gag–specific CD8⁺ T-cell responses elicited by SIV ΔgL vs. gL-intact 68–1 RhCMV/SIV vaccine vectors. (A) Box plots compare the memory differentiation phenotype of CD4⁺ and CD8⁺ T cells in blood of group 1 (ΔgL 68–1 RhCMV/SIV vaccinated) vs. group 2 (gL-intact 68–1 RhCMV/SIV vaccinated) RM responding to the Gag 15-mer peptide mix (TNF-α and/or IFN-γ) during plateau phase (weeks 39 to 42 post–initial vaccination). Memory differentiation state (% of each subset within the total response) based on CD28 and CCR7 delineates central memory (T_CM), transitional effector-memory (T_TrEM), and effector-memory (T_EM). The 2-sample WRS test (adjusted for multiple comparisons) was used to determine the significance of differences between group 1 (ΔgL) vs. group 2 (gL-intact) 68–1 RhCMV/SIV vaccination for the percentage of Gag-specific T-cells within each memory subset (Holm-adjusted P-values ≤0.05 are shown). (B) Box plots compare the frequency of CD4⁺ and CD8⁺ memory T-cells in blood responding to the Gag 15-mer peptide mix with TNF-α, IFN-γ, IL-2, and MIP-1α production, alone and in all combinations (same samples as shown in A). Results were grouped according to the number of cytokines secreted. The 2-sample WRS test (adjusted for multiple comparisons) was used to determine the significance of differences in the percentage of Gag-specific T cells expressing different combinations of 1, 2, 3, or 4 cytokines between group 1 and group 2 (Holm-adjusted P-values ≤0.05 are shown).

Figure 5.

Efficacy analysis of ΔgL vs. gL-intact 68–1 RhCMV/SIV vaccines upon SIV challenge. (A) Assessment of the outcome of SIV infection after repeated, limiting-dose intrarectal SIVmac239 challenge of group 1 (ΔgL 68–1) vs. group 2 (gL-intact 68–1) RMs by longitudinal analysis of plasma viral load (left panels) and of the de novo development of SIV Vif-specific CD4⁺ and CD8⁺ T-cell responses (middle and right panels, respectively). RMs were SIV challenged until the onset of sustained viremia and/or above-threshold SIV Vif-specific (and SIV Env-specific, See Fig. S4) T-cell responses. The SIV dose administered 1 to 3 weeks prior to the initial detection of viremia and/or Vif-response detection was considered the infecting challenge (week 0). RMs with sustained viremia were considered nonprotected (black); RMs with no or transient viremia but demonstrating sustained above-threshold SIV Vif- (and Env-) specific T-cell responses were considered protected (red). The Fisher exact test indicated that the difference in the proportion of protected RMs after ΔgL (group 1) vs. gL-intact (group 2) 68–1 vaccination was not significant. (B) Controlled SIV infection was confirmed in 3 of the aviremic, protected ΔgL 68–1 RhCMV/SIV–vaccinated RM by demonstration of sustained plasma SIV viremia in SIV-naïve recipient RMs after adoptive transfer of bone marrow and peripheral lymph node cells collected between days 28 and 56 post–the infecting challenge (top left table and middle figure). In contrast, adoptive transfer of infection was not observed in cells taken from the same RM 784 days post–effective challenge (bottom left table), indicating subsequent clearance of the controlled infection (right panel).