Co-immunization of DNA and Protein in the Same Anatomical Sites Induces Superior Protective Immune Responses against SHIV Challenge

Abstract

We compare immunogenicity and protective efficacy of an HIV vaccine comprised of env and gag DNA and Env (Envelope) proteins by co-administration of the vaccine components in the same muscles or by separate administration of DNA + protein in contralateral sites in female rhesus macaques. The 6-valent vaccine includes gp145 Env DNAs, representing six sequentially isolated Envs from the HIV-infected individual CH505, and matching GLA-SE-adjuvanted gp120 Env proteins. Interestingly, only macaques in the co-administration vaccine group are protected against SHIV CH505 acquisition after repeated low-dose intravaginal challenge and show 67% risk reduction per exposure. Macaques in the co-administration group develop higher Env-specific humoral and cellular immune responses. Non-neutralizing Env antibodies, ADCC, and antibodies binding to FcγRIIIa are associated with decreased transmission risk. These data suggest that simultaneous recognition, processing, and presentation of DNA + Env protein in the same draining lymph nodes play a critical role in the development of protective immunity.

Keywords: ADCC; DNA; FcγR; HIV; Indian rhesus macaques; antibody; cellular immunity; humoral immunity; protein; vaccine.

Figure 1.. Co-administration Group Shows Significant Protection from SHIV.CH505 Infection

(A) Schematic representation of vaccine delivery of the two components (DNA and protein) in the two vaccination regimens, “Co-administration” in the same anatomical sites and “separate administration” in contralateral sites. The co-administration group received DNA delivered via IM/EP followed immediately by IM injection of the adjuvanted protein. The separate administration group received the vaccine components in different anatomical sites with DNA delivered by IM/EP in the left site and protein IM in the right site. The same vaccine components and the same total vaccine dose were used for both regimens. (B) Vaccination schedule indicating the sequentially isolated CH505 immunogens used. Five weeks after the last vaccination, the animals were exposed weekly to repeated low-dose vaginal challenges using SHIV.CH505. (C) Kaplan-Meier curves show the viral acquisition rate after repeated low-dose SHIV.CH505 challenges of the two vaccine groups (n = 18 and 17, respectively) and the control group (n = 20). The RMs were exposed to 15 weekly intravaginal challenges. Infection was defined by two consecutive positive plasma VL measurements. No RMs were censored. p value, exact log-rank test.

Figure 2.. Co-administration Regimen Induced Higher Env Immune Responses

(A) Kinetics of vaccine-induced CH505.M11 Env antibodies (Abs) were measured by ELISA and Env Ab titers of individual RMs are shown as area under the curve (AUC) at the time of vaccination and 2 weeks later. Red symbols, co-administration group; blue symbols, separate administration vaccine group. Red numbers indicate p values showing significant higher values measured in the co-administration group. (B) Summary of the differences in Env Ab titers measured after each vaccination against a panel of CH505 gp120 proteins. p values denote statistically significant higher values in the co-administration group; p values (p = 0.051- < 0.06) indicate trend; ns, not significant. ND, not determined. (C–F) Comparison of HIV Env and SIV Gag Abs and T cell responses between the vaccine groups after the last vaccination. (C and D) Ab responses measured to (C) HIV CH505.TF gp120 Env and (D) SIV Gag. (E and F) T cell responses (% antigen-specific IFN-γ⁺) measured with peptides covering (E) all CH505 gp145 Env proteins used in the vaccine (F) SIV p57^Gag. Similar data were obtained from samples analyzed after fifth vaccination. (G and H) Antigen-specific IFN-γ⁺ T cells are presented as a percentange of CD4⁺ (G) and CD8⁺ (H) T cell subsets. (I and J) Direct correlation between the CH505 Env-specific CM-like CD4⁺ IFN-γ⁺ T cells (presented as a percentage of the CD3⁺ cells) and (I) CH505.TF Ab levels after sixth vaccination and (J) summary of correlations with different Env Abs after fifth vaccination. R and p values (Spearman) are given. (K) Abs to CH505.M5 Env were measured in vaginal secretion after vaccination 6. Values are presented as specific binding activity (SA) and were calculated as mean fluorescence intensity (MFI) 3 dilution/total IgG concentration (micrograms per milliliter). p values (Mann-Whitney test) are given for all comparisons.

Figure 3.. Non-neutralizing Abs Contribute to Reduced Risk of Infection

(A and B) Direct correlation of CH505 bAb titers (AUC, log₁₀) and the number of SHIV.CH505 exposures to infection plotted using plasma after (A) third vaccination and (B) summarized for all Env antigens after third and fourth vaccinations. No correlations were found after the fifth and sixth vaccinations. p values (Spearman) are listed, p values in parentheses indicate trend. (C) NAbs to tier-1A CH505.w4.3 (upper panel) and the tier-2 CH505.TF (lower panel) measured in serum 2 weeks after the third to sixth vaccinations. Black dotted line indicates the limit of detection (LOD) of neutralization in the TZM-bl assay. CH505.w4.3 differs from the tier-2 CH505.TF by a single point mutation (W663G), located in the MPER (membrane-proximal extracellular region). (D) Binding of Env-Abs exposed on the surface of HIV.CH505 infected cells measured by the infected cell Ab binding assay (ICABA) showing MFI of the two vaccine groups. p values are from Wilcoxon test (statistical analysis software). (E and F) Correlation between ADCC peak (maximum % GrzB activity) and number of SHIV CH505 exposures to infection (E) after third vaccination and (F) summary of correlations measured after third and fourth vaccinations. No correlations were found with the peak ADCC after fifth and sixth vaccinations. p values (Spearman) are given. (G) Binding of CH505-Env-specific Abs to FcγRIIIa1. (H and I) Direct correlations between ADCC titers and CH505.TF-specific Ab binding to FcγRIIIa1 and FcγRIIIa3 showing correlation between (H) CH505.TF gp120 Abs and FcγRIIIa1 and (I) different CH505 Abs binding to FcγRIIIa3 after sixth vaccination. Spearman R and p values are given.

Figure 4.. Immunogenicity Profiles Robustly Distinguishes Animals Vaccinated with the Co-administration and Separate Administration Vaccine Regimens

(A) Balanced accuracy of models learned on training data when applied to test data across 100 replicates of 10-fold cross validation. Models learned from actual data (left) perform substantially better (Cliff’s delta, 1.0) than those learned from permuted data (right). Average accuracy across cross-validation runs is reported in inset. (B) Confusion matrix depicting the proportion of animals in each study arm predicted correctly/incorrectly. (C) Model confidence, defined by probability of belonging to the separate side class. Dotted line indicates the decision boundary. (D) Feature contributions to the simplified final model. (E) Principal-component (PC) biplot of features contributing to the simplified final model. Simplified model accuracy is reported in inset. Animals are represented as dots, with color indicating vaccine arm. Classification performance over time could not be strictly compared, as different immunogenicity tests were performed at different time points, and in some cases, samples from all macaques were not available for all time points for all tests (Table S2). Simplified forms of the final models learned on one time point were applied to data from other time points and demonstrated good consistency in defining signatures of group-specific immunogenicity that were robust across longitudinal time points during the immunizations (Table S5). The data from the analysis performed after third vaccination showing peak accuracy is shown boxed.

Figure 5.. Immune Response Profiles Robustly Distinguish Resistant Animals in the Co-administration Group

(A) Histogram of challenge outcomes for all RMs with sensitive and resistant animals defined as those infected (purple box) or not infected (green box) prior to the 10th challenge. (B–F) Models of infection resistance learned from immunogenicity data post-third immunization for the co-administration group. A classifier was trained to use immunogenicity data to distinguish between animals based on challenge resistance. (B) Balanced accuracy of models learned on training data when applied to test data. Models learned from actual data (left) perform substantially better (Cliff’s delta, 0.84) than those learned from permuted data (right). Robust performance and good overall average accuracy (83%) in the setting of repeated cross validation was achieved. (C) Confusion matrix depicting the proportion of animals in each study arm predicted correctly/incorrectly. (D) Model confidence defined by probability of belonging to the assigned class. (E and F) A final model was trained on the complete data and simplified to a sparse, two-feature model. (E) Coefficients of features in the simplified final model. (F) PC biplot of features contributing to the simplified final model with accuracy reported in inset. Resistant RMs are indicated in green; susceptible RMs are indicated in purple.

Figure 6.. CH505-Specific Ab Binding to FcγRIIIa Robustly Predicts Challenge Resistance after the Final Vaccination

(A) Balanced accuracy of models learned on training data when applied to test data. Models learned from actual data (left) perform moderately better (Cliff’s delta, 0.33) than those learned from permuted data (right). Average accuracy across cross-validation runs is reported in inset. (B and C) Confusion matrix confidence (B) and model confidence (C) as described in Figure 5. (D and E) Coefficients of features (D) and PC biplot of features (D) contributing to the final model with accuracy reported in inset. Resistant RMs are indicated in green; susceptible RMs are indicated in purple.

Figure 7.. Both Vaccine Regimens Induce Immune Responses Able to Reduce Viremia

(A) VL of all infected animals monitored over 18 weeks of follow-up showing the geometric mean VL. The p values (Kruskal-Wallis tests of differences over the three groups at the individual times) weeks 2–9 are given, with wide fluctuations in some of the individual levels at weeks 6–9. (B) Comparison of early viremia (weeks 0–4) as AUC with p values of pairwise comparisons (Kruskal-Wallis test). (C) Changes in the level of virus-specific (Env and Gag) CD8⁺ T cells measured after sixth vaccination (3 weeks before challenge start) and 4 weeks after infection (expressed as a percentage of total CD3⁺ T cells). p value, Wilcoxon signed rank test.