Vaccine-Induced Protection from Homologous Tier 2 SHIV Challenge in Nonhuman Primates Depends on Serum-Neutralizing Antibody Titers

Abstract

Passive administration of HIV neutralizing antibodies (nAbs) can protect macaques from hard-to-neutralize (tier 2) chimeric simian-human immunodeficiency virus (SHIV) challenge. However, conditions for nAb-mediated protection after vaccination have not been established. Here, we selected groups of 6 rhesus macaques with either high or low serum nAb titers from a total of 78 animals immunized with recombinant native-like (SOSIP) Env trimers. Repeat intrarectal challenge with homologous tier 2 SHIV_BG505 led to rapid infection in unimmunized and low-titer animals. High-titer animals, however, demonstrated protection that was gradually lost as nAb titers waned over time. An autologous serum ID₅₀ nAb titer of ∼1:500 afforded more than 90% protection from medium-dose SHIV infection. In contrast, antibody-dependent cellular cytotoxicity and T cell activity did not correlate with protection. Therefore, Env protein-based vaccination strategies can protect against hard-to-neutralize SHIV challenge in rhesus macaques by inducing tier 2 nAbs, provided appropriate neutralizing titers can be reached and maintained.

Keywords: ADCC; BG505; HIV vaccine; correlates of protection; neutralizing antibodies; non-human primates; tier 2 protection; vaccination.

Graphical abstract

Figure 1

High and Low nAb Titer Group Animals Have Significantly Different Serum nAb Titers after Env Trimer Immunization (A) Animals, except for the controls, received a booster immunization using the same immunogen that had last been used during the preceding immunization study (Pauthner et al., 2017), typically 100 μg SOSIP trimer adjuvanted in a soluble ISCOMs-class saponin. Intrarectal (IR) challenges with SHIV_BG505 S375Y commenced 4 weeks thereafter. All groups of animals received six IR challenges starting at week 0. High nAb titer animals that had undetectable serum viral loads at week 6 received a second set of 6 weekly IR challenges starting week 11. (B and C) Serum neutralizing ID₅₀ titers in high and low nAb titer animals at week −2: BG505 S375Y pseudovirus (B) and rhCD4⁺ T-cell-grown SHIV_BG505 S375Y challenge stock. (C) Shown are geometric mean titers with geometric standard deviations, significant differences were determined using two-tailed Mann-Whitney U tests. See also Figure S1.

Figure 2

High nAb Titer Animals Show Robust Protection from SHIV_BG505 Challenge (A–C) Viral loads of animals throughout the challenge schedule: unimmunized concurrent controls (A), low nAb titer (B), and high nAb titer (C) macaques. IR challenges are indicated with vertical dotted lines. Horizontal dotted lines denote the limit of detection. (D) Kaplan-Meier curves indicating percent uninfected animals over the duration of the study. Challenge time points are indicated with red arrows. Significance levels are indicated with stars; ^∗∗∗∗p < 0.0001. Statistics were calculated for both the first (dotted line at week 8) and second challenge sets (see Table S3). (E) Geometric mean viral loads of indicated groups, normalized to the detection of viremia in the blood. Horizontal lines at 10⁵ and 10⁷ viral RNA copies/mL serve as visual aids. (F) Comparison of peak viral loads between high nAb titer (High), low nAb titer (Low), and unimmunized (Ctrl) animals. Geometric mean viral loads are shown with geometric standard deviations. Significant differences were determined using two-tailed Mann-Whitney U tests. See also Figure S2.

Figure 3

Autologous Tier 2 nAb Titers Surge in Vaccinated Animals after Infection Serum neutralizing antibody titers throughout the challenge schedule: BG505 S375Y pseudovirus ID₅₀ nAb titers rise 8–12 weeks after infection in unimmunized animals (A) or 1–2 weeks after detection of viremia in low nAb titer animals (B). BG505 S375Y pseudovirus ID₅₀ nAb titers in macaques that became infected over time (C) or showed sterilizing protection (D). First detection of plasma viremia is indicated by colored arrows corresponding to the animal IDs shown in the respective figure legends. See also Figure S3.

Figure 4

Viral Escape Follows Resurgence of nAb Titers after Infection (A and B) Viral sequencing after SHIV challenge and infection. PacBio viral sequencing data of two high nAb titer animals, 12-143 (A) and 12-137 (B), indicates putative escape mutations after infection, at the time points indicated. Challenge denotes challenge stock. (C) Negative stain electron micrograph of a C3-V5 directed serum specificity (purple) observed among cleaved serum Fab fragments of animal 12-137 at week 0 bound to BG505 SOSIP.664. Shown are a top view (upper panel) and a 90° rotated front view (lower panel). N355 is highlighted in red. (D and E) Serum neutralization data of animals 12-143 (D) and 12-137 (E) at indicated time points against either BG505 S375Y pseudovirus (WT) or indicated mutants thereof. Error bars indicate technical replicates.

Figure 5

Protection Is Associated with Serum nAb Titers Greater than ∼1:500 (A) Correlation of BG505 S375Y pseudovirus ID₅₀ nAb titers at week −2 with the number of weeks until viremia was detected. (B) BG505 S375Y pseudovirus ID₅₀ nAb titers of control (Ctrl.), low nAb titer (Low), and high nAb titer (High) animals 7 days before detection of viral load in the blood and at week 20 for protected (Prot.) animals that showed sterilizing protection throughout the study. All nAb titers were measured in TZM-bl assays. Shown are geometric mean titers with geometric standard deviations. Correlations were calculated using Spearman correlation tests, comparisons between groups were calculated using Mann-Whitney U tests. Horizontal lines indicate 50% and 90% protective nAb titers as defined in (C). (C) The 5%, median, and 95% credible intervals (CI) are shown for the probability of infection in relation to serum BG505 S375Y pseudovirus nAb titer, inferred using a modified Bayesian logistic regression model (see Figure S5). The posterior median infection probability at the limit of nAb titer detection was 77%, corresponding to an AID₇₇. A median infection probability of 50% is attained with an ID₅₀ titer of 1:90 (red line, CI: 34-178), and an infection probability of 10% with an ID₅₀ titer of 1:476 (blue line, CI: 272-991). See also Figures S4 and S5.

Figure 6

HIV Env-Specific CD4⁺ T Cells and Env-Specific CD8⁺ T Cells at Week 0 Are Not Associated with the Observed Protection from Infection (A–C) Representative flow plots of Env-specific CD4⁺ T cells from week 0 PBMCs: using an OX40/4-1BB AIM assay (Reiss et al., 2017) (A), intracellular staining (ICS) assay for IFN-γ (B), and ICS assay for TNF/CD40L (C) when not stimulated (NS) versus stimulated with antigen (Env). (D) Representative flow plot of IFN-γ and TNF expression in CD8⁺ T cells by ICS when not stimulated (NS) versus stimulated with antigen (Env). (E–G) Quantification of the percent of CD4⁺ T cells that are Env-specific based on: OX40/4-1BB (E), IFN-γ (F), or CD40L/TNF (G) expression. (H) Quantification of the percent of CD8⁺ T cells that are Env-specific based on IFN-γ and TNF expression. Signal from the unstimulated condition was subtracted from the antigen-specific signal for each sample. Each dot represents an individual animal. Shown are high and low nAb titer, as well as control group animals. See also Figure S6.

Figure 7

ADCC Activity at Week 0 Measured in SHIV Infection as well as gp120-Based Assays Is Not Associated with the Observed Protection from Infection (A–D) ADCC activity from sera of high and low nAb titer as well as control animals at week 0. ADCC activity in titrated sera was measured using SHIV_BG505 challenge stock infected CEM.NKR luciferase-reporter target cells and CD16 transfected KHYG-1 effector cells (A) or in 1:250 diluted sera by flow cytometric analysis of ADCC activity in either p27⁺ SHIV_BG505-infected CEM.NKR cells (B) or BG505 g120-coated CEM.NKR cells (C), using PBMCs as effector cells. Shown are means with standard deviations. ADCC activity in BG505 gp120-coated CEM.NKR cells correlated with BG505 gp120 binding titers (D). (E–G) ELISA EC₅₀ binding titers at week 0 to: BG505 SOSIP.664 (E), BG505 V3-peptide (F), or BG505 gp120 (G). Sera from high and low nAb titer animals, as well as unimmunized control animals were tested for ELISA binding titers at week 0. Shown are geometric mean titers with geometric standard deviations. Correlations were calculated using Spearman correlation tests, comparisons between groups were calculated using two-tailed Mann-Whitney U tests. See also Figure S7.