Heterologous prime-boost regimens using rAd35 and rMVA vectors elicit stronger cellular immune responses to HIV proteins than homologous regimens

Abstract

We characterized prime-boost vaccine regimens using heterologous and homologous vector and gene inserts. Heterologous regimens offer a promising approach that focuses the cell-mediated immune response on the insert and away from vector-dominated responses. Ad35-GRIN/ENV (Ad35-GE) vaccine is comprised of two vectors containing sequences from HIV-1 subtype A gag, rt, int, nef (Ad35-GRIN) and env (Ad35-ENV). MVA-CMDR (MVA-C), MVA-KEA (MVA-K) and MVA-TZC (MVA-T) vaccines contain gag, env and pol genes from HIV-1 subtypes CRF01_AE, A and C, respectively. Balb/c mice were immunized with different heterologous and homologous vector and insert prime-boost combinations. HIV and vector-specific immune responses were quantified post-boost vaccination. Gag-specific IFN-γ ELISPOT, intracellular cytokine staining (ICS) (CD107a, IFN-γ, TNF-α and IL-2), pentamer staining and T-cell phenotyping were used to differentiate responses to inserts and vectors. Ad35-GE prime followed by boost with any of the recombinant MVA constructs (rMVA) induced CD8+ Gag-specific responses superior to Ad35-GE-Ad35-GE or rMVA-rMVA prime-boost combinations. Notably, there was a shift toward insert-focus responses using heterologous vector prime-boost regimens. Gag-specific central and effector memory T cells were generated more rapidly and in greater numbers in the heterologous compared to the homologous prime-boost regimens. These results suggest that heterologous prime-boost vaccination regimens enhance immunity by increasing the magnitude, onset and multifunctionality of the insert-specific cell-mediated immune response compared to homologous vaccination regimens. This study supports the rationale for testing heterologous prime-boost regimens in humans.

Figure 1. Schematic of the two studies.

Study 1 tested 4 vaccine combinations (5 mice/group) plus 10 naïve mice. Spleens and serum were collected at day 14 after the boost corresponding to day 35 on the schematic. Study 2 included the set of pairings from the first study (highlighted in red) in addition to the pairings indicated. Each group included 8 BALB/c mice that were vaccinated at days 0 and 21 as indicated. In this study, serum and blood were also harvested at the time indicated in the schematic. rMVA and rAd35 constructs were injected intramuscularly at 1×10⁷ pfu and 2×10¹⁰ vp, respectively.

Figure 2. CD8+ ICS assays: CD8+ T-cell functional responses to the A) Gag immunodominant variant KDTI and B) the immunodominant 050 MVA peptide.

The responses are shown as the mean plus Standard Deviation calculated by combining the results from both studies. Vaccination regimens are on the X-axis, MVA = MVA-C and Ad35 = Ad35-GE; horizontal color coded bars show which cytokine or degranulation marker reached a statistically significant increase (p<0.05). Kruskall-Wallis test and Dunn's Multiple Comparison test were applied. Only significant results are shown.

Figure 3. IFN-γ ELISPOT results from Study 1.

Graphs represent means SFU/10⁶ cells + standard deviations of the responses to A) Gag peptide pools and B) envelope peptide pools. CMDR pools match the MVA-C vaccine inserts sequences, P1 peptide pools match the Ad35-GE inserts sequences. GAG A VIP is a peptide pool designed to maximize coverage of the most frequent 10-mer epitope variants in the HIV subtype A epidemic. Vaccination regimens are on the X-axis: MVA = MVA-C and Ad35 = Ad35-GE, horizontal bars show which peptide pool response reached a statistically significant increase. Kruskall-Wallis test and Dunn's Multiple Comparison test were applied. Only significant results are shown. ∧ symbol identify responses to peptide pools that were statistically significant different (p<0.05 Mann Whitney test) within the same regimen.

Figure 4. CD8+ T cells with different pentamer binding capacity are elicited by the heterologous prime-boost regimens.

Depicted are the density plots of CD8+ T cell elicited by the indicated vaccine regimens and their ability to bind to the pentamer-bound Gag immunodominant motifs. The upper right quadrant represent T cell that can cross-react with both immunodominant motifs.

Figure 5. Memory phenotype of CD8+ T cells collected sequentially from mice immunized with the indicated regimens.

Density plot in gray represent the memory phenotype of the cells, superimposed red dots represent the KDTI pentamer positive CD8+ T cells. Gates that define the memory subpopulations are only shown in the upper left graph, the percentage of the different memory subpopulation defined by these gates are expressed in the upper right corner in each graph. CM: central memory cells, Effector: effector cells, EM: effector memory cells.

Figure 6. Day 7 post-boost representation of pentamer positive lymphocytes.

Percentage of pentamer-binding positive lymphocytes are grouped on the x-axis according to their memory phenotype. Panels A and C show the KDTI pentamer-binding positive and panels B and D the KETI positive. Panels A and B show the phenotype of the MVA homologous regimens and panels C and D of the heterologous regimens.