HIV-1 Gag-specific immunity induced by a lentivector-based vaccine directed to dendritic cells

Abstract

Lentivectors (LVs) have attracted considerable interest for their potential as a vaccine delivery vehicle. In this study, we evaluate in mice a dendritic cell (DC)-directed LV system encoding the Gag protein of human immunodeficiency virus (HIV) (LV-Gag) as a potential vaccine for inducing an anti-HIV immune response. The DC-directed specificity is achieved through pseudotyping the vector with an engineered Sindbis virus glycoprotein capable of selectively binding to the DC-SIGN protein. A single immunization by this vector induces a durable HIV Gag-specific immune response. We investigated the antigen-specific immunity and T-cell memory generated by a prime/boost vaccine regimen delivered by either successive LV-Gag injections or a DNA prime/LV-Gag boost protocol. We found that both prime/boost regimens significantly enhance cellular and humoral immune responses. Importantly, a heterologous DNA prime/LV-Gag boost regimen results in superior Gag-specific T-cell responses as compared with a DNA prime/adenovector boost immunization. It induces not only a higher magnitude response, as measured by Gag-specific tetramer analysis and intracellular IFN-gamma staining, but also a better quality of response evidenced by a wider mix of cytokines produced by the Gag-specific CD8(+) and CD4(+) T cells. A boosting immunization with LV-Gag also generates T cells reactive to a broader range of Gag-derived epitopes. These results demonstrate that this DC-directed LV immunization is a potent modality for eliciting anti-HIV immune responses.

Fig. 1.

Comparison of immune responses generated from different injection routes after a single immunization. (A) Schematic representation of a lentiviral backbone construct encoding the full sequence of a HIV-1 subtype B Gag antigen. R, U5, and ΔU3 are components of the long terminal repeat (LTR) and ΔU3 contains the self-inactivating deletion; SD: splicing donor; SA: splicing acceptor; ψ and ΔGag: the encapsulation sequence; RRE: the Rev-responsive element; Ubi: human ubiquitin-C promoter; WPRE: woodchuck hepatitis virus posttranductional regulatory element. (B) Five groups of BALB/c mice were immunized with 5 × 10⁶ TU (Transduction Units) of LV-Gag by a s.c., footpad (f.p.), intramuscular (i.m.), i.p., or intradermal (i.d.) injection route. Two weeks postimmunization, spleen cells were harvested and analyzed for the frequency of Gag-specific CD8⁺ T cells by H2-K_d-AMQMLKETI-PE tetramer and CD44 staining. Spleen cells were also restimulated in vitro with the HIV-1 Gag peptide (AMQMLKETI). Intracellular cytokine staining (ICCS) was performed to assess the IFN-γ response. (C) Sera from different groups of mice were harvested 2 weeks postimmunization. IgG and IgM antibody responses against HIV-1 Gag were detected by ELISA. Each group consisted of three mice.

Fig. 2.

DC-directed LV can effectively boost HIV-1 Gag-specific immune response. (A) BALB/c mice were immunized with PBS (●), BMDCs (1 × 10⁶⁾ loaded by the HIV-1 Gag peptide (AMQMLKETI) (■), empty LV lacking the Gag transgene (5 × 10⁶ TU) (▲), or LV-Gag (5 × 10⁶ TU) (▼). The immune responses of spleen cells upon restimulation with the Gag dominant peptide were estimated by IFN-γ ICCS 2 weeks postinjection (*, P < 0.01; **, P < 0.001) (B–D) Four vaccine groups received PBS, single immunization of LV-Gag (LV-Gag once), LV-Gag prime/LV-Gag boost (LV-Gag/LV-Gag), or DNA prime/LV-Gag boost (DNA/LV-Gag). Splenocytes from vaccinated animals were analyzed for Gag-specific response by H2-K_d-AMQMLKETI-PE tetramer staining (B), IFN-γ ICCS (C), and mouse serum ELISAs for IgG and IgM (D). The data shown are mean values of triplicates ± SD.

Fig. 3.

Comparison of magnitude, kinetics and memory responses of Gag-specific CD8⁺ T cells after immunization with LV-Gag and rAd5-Gag. Six groups of BALB/c mice received the following vaccination regimens: PBS, LV-Gag (LV, 5 × 10⁶ TU), rAd5-Gag (rAd5, 10¹⁰ VP), LV-Gag prime/LV-Gag boost (LV/LV), DNA prime/LV-Gag boost (DNA/LV), and DNA prime/rAd5-Gag boost (DNA/rAd5). Vaccine-induced HIV Gag-specific immune responses were analyzed by: (A) percentage of IFN-γ- or Gag-tetramer-positive CD8⁺ T cells (*, P < 0.05; **, P < 0.005); (B) kinetics of the total frequency of IFN-γ-producing CD8⁺ T cells of LV-Gag and rAd5-Gag groups on indicated time points after immunization; and (C) division of central memory (T_CM, CD44^highCD62L⁺) and effector memory (T_EM, CD44^highCD62L⁻) CD8⁺ T cells of LV-Gag (LV) and rAd5-Gag (rAd5) groups by surface staining.

Fig. 4.

Generation of multifunctional CD4⁺ and CD8⁺ responses by prime/boost immunization regimens. Splenocytes of DNA/rAd5-Gag (DNA/rAd5), LV-Gag/LV-Gag (LV/LV) and DNA/LV-Gag (DNA/LV) groups of BALB/c mice were stimulated with the pooled HIV-1 Gag peptides (2.5 μg/mL for each peptide) for 6 h, and analyzed by an eight-color ICCS assay to assess: (A) the fraction of total responding CD4⁺ or CD8⁺ T cells expressing each of the seven possible combinations of IFN-γ, IL-2, and TNF-α; and (B) the frequency and proportion of responding CD4⁺ T cells expressing all three cytokines (IFN-γ⁺IL-2⁺TNF-α⁺: γ+2+α+), two cytokines (IFN-γ⁻IL-2⁺TNF-α⁺: γ−2+α+; or IFN-γ⁺IL-2⁺TNF-α⁻: γ+2+α−), or one cytokine (IFN-γ⁻IL-2⁺TNF-α⁻: γ−2+α−).

Fig. 5.

Breadth of HIV-1 Gag-specific responses to LV-Gag-based vaccination. (A) A library of 123 15-mer peptides spanning the entire HIV-1 subtype B Gag sequence was divided into 23 pools (P1–P23) as indicated by the peptide matrix table. (B and C) Spleen cells of DNA/rAd5-Gag, LV-Gag/LV-Gag, and DNA/LV-Gag groups of BALB/c mice were harvested, stimulated with one of peptide pools for 18–24 h, and assayed by IFN-γ ELISPOT. Each group consisted of three mice. Number of highly reactive peptide pools versus number of nonreactive peptide pools for each group of mice was summarized as shown in (B). The threshold for defining a nonreactive peptide pool was based on the ELISPOT readout comparable with the control PBS group. A peptide pool was defined as highly reactive when its stimulated response was 5 times higher than that of the control PBS reading and could be obviously detected by an ICCS assay.