Priming with a Potent HIV-1 DNA Vaccine Frames the Quality of Immune Responses prior to a Poxvirus and Protein Boost

Abstract

The use of heterologous immunization regimens and improved vector systems has led to increases in immunogenicity of HIV-1 vaccine candidates in nonhuman primates. In order to resolve interrelations between different delivery modalities, three different poxvirus boost regimens were compared. Three groups of rhesus macaques were each primed with the same DNA vaccine encoding Gag, Pol, Nef, and gp140. The groups were then boosted with either the vaccinia virus strain NYVAC or a variant with improved replication competence in human cells, termed NYVAC-KC. The latter was administered either by scarification or intramuscularly. Finally, macaques were boosted with adjuvanted gp120 protein to enhance humoral responses. The regimen elicited very potent CD4⁺ and CD8⁺ T cell responses in a well-balanced manner, peaking 2 weeks after the boost. T cells were broadly reactive and polyfunctional. All animals exhibited antigen-specific humoral responses already after the poxvirus boost, which further increased following protein administration. Polyclonal reactivity of IgG antibodies was highest against HIV-1 clade C Env proteins, with considerable cross-reactivity to other clades. Substantial effector functional activities (antibody-dependent cell-mediated cytotoxicity and antibody-dependent cell-mediated virus inhibition) were observed in serum obtained after the last protein boost. Notably, major differences between the groups were absent, indicating that the potent priming induced by the DNA vaccine initially framed the immune responses in such a way that the subsequent boosts with NYVAC and protein led only to an increase in the response magnitudes without skewing the quality. This study highlights the importance of selecting the best combination of vector systems in heterologous prime-boost vaccination regimens.IMPORTANCE The evaluation of HIV vaccine efficacy trials indicates that protection would most likely correlate with a polyfunctional immune response involving several effector functions from all arms of the immune system. Heterologous prime-boost regimens have been shown to elicit vigorous T cell and antibody responses in nonhuman primates that, however, qualitatively and quantitatively differ depending on the respective vector systems used. The present study evaluated a DNA prime and poxvirus and protein boost regimen and compared how two poxvirus vectors with various degrees of replication capacity and two different delivery modalities-conventional intramuscular delivery and percutaneous delivery by scarification-impact several immune effectors. It was found that despite the different poxvirus boosts, the overall immune responses in the three groups were similar, suggesting the potent DNA priming as the major determining factor of immune responses. These findings emphasize the importance of selecting optimal priming agents in heterologous prime-boost vaccination settings.

Keywords: DNA vaccine; Gag-Pol-Nef; NYVAC; NYVAC-KC; T cell responses; antibody responses; gp140; human immunodeficiency virus; nonhuman primates; vaccine.

FIG 1

Immunization schedule for the DNA prime and NYVAC and protein boost regimen. Three groups of 8 macaques each were immunized three times with DNA, followed by a single immunization with an NYVAC vector and three immunizations with protein. NYVAC-C-KC (enhanced replication competence) was administered by scarification in group A, and by i.m. injection in group B. In group C, NYVAC-C (restricted replication competence) was administered by i.m. injection. Blood was collected for ELISpot analysis, intracellular cytokine staining (ICS), or antibody analysis at the indicated time points.

FIG 2

The DNA prime and NYVAC boost regimen elicits vigorous HIV-1-specific T cell responses. PBMCs were freshly isolated at the indicated time points and restimulated in parallel with nine different peptide pools in an ELISpot assay. The numbers of IFN-γ-producing cells per million PBMCs were enumerated as spot-forming units (SFUs) and summed for all the peptide pools. Results are shown per animal, along with the median and interquartile range per group.

FIG 3

The combined immunizations elicited balanced, polyfunctional and broad CD4⁺ and CD8⁺ T cell responses to HIV-1 antigens. PBMCs obtained at the indicated time points (weeks after the first immunization) were stimulated with the nine different peptide pools and stained for CD3, CD4, and CD8 and intracellular IFN-γ, IL-2, and TNF. Responding CD4⁺ and CD8⁺ T cells were quantified by flow cytometry, as indicated. (A) Fractions of restimulated CD4⁺ and CD8⁺ T cells summed for all the peptide pools and any cytokine are shown with median and interquartile ranges. Significant differences between time points in a paired Wilcoxon signed rank test are marked with an asterisk (P < 0.05, Bonferroni corrected). (B) Median values of the summed fractions of T cells restimulated with any peptide pool, broken down according to the cytokines produced by the cells. (C) Median values of the summed fractions of T cells secreting any combination of the three cytokines, broken down according to the peptide pools used.

FIG 4

The DNA/NYVAC/protein immunization course elicited substantial IgG antibody responses and negligible IgA antibody responses to HIV-1 antigens. (A and B) IgG antibodies from serum dilutions binding to the indicated readout antigens coated on beads were quantified using a customized Luminex assay. The magnitudes of the antibody binding responses are given as values of the area under the titration curve (AUC), and medians and interquartile ranges are shown. (A) Representative time courses of responses toward the indicated antigens. n.d., not done. (B) Comparative responses of the week 36 sera toward all antigens assessed. (C) IgA antibodies from the lowest serum dilution binding to the indicated readout antigens were quantified using the Luminex assay. The magnitudes are given as mean fluorescence intensity (MFI) values, and medians and interquartile ranges are shown.

FIG 5

High-titer V1V2-directed responses were elicited after the protein boosts. IgG antibodies directed toward an MLV gp70-scaffolded V1V2 peptide from isolate Case-A2 were measured by ELISA. The reciprocal titer matching an optical density (OD) threshold value of 0.1 interpolated from the titration curve is given for each animal along with group medians and interquartile ranges.

FIG 6

Macaque sera had low neutralization activity in the TZM-bl assay. Serum dilutions were tested for neutralization of pseudoviruses, and the reciprocal dilution at which 50% inhibition (ID₅₀) was observed is given for each animal, with medians and interquartile ranges. (A) Time courses of neutralization of the pseudoviruses carrying MW965.26 or MN.3. (B) ID₅₀ values of the week 51 serum samples for pseudoviruses carrying Envs from the indicated isolates.

FIG 7

Sera obtained from macaques at week 51 showed vigorous ADCC as well as ADCVI activity. (A) CEM.NKR_CCR5 target cells coated with TV1 gp120 were coincubated with PBMC preparations (including NK cells) at an effector/target cell ratio of 30:1 in the presence of plasma dilutions from the immunized macaques. The amount of granzyme B released by NK cells was measured using a fluorescent substrate. The plasma dilution needed to match the granzyme B activity cutoff value was calculated, and the reciprocal is given as the titer for values above the assay cutoff of 100, along with median and interquartile ranges. (B) CEM.NKR_CCR5 cells were infected with infectious HIV-1 strains DU156 and DU422 and then coincubated with PBMC effector cells in the presence of plasma at an effector/target cell ratio of 10:1. The amount of virus released was measured by a p24 ELISA, and the inhibition of virus production compared with that in samples lacking plasma is given.