Intrastructural Help: Harnessing T Helper Cells Induced by Licensed Vaccines for Improvement of HIV Env Antibody Responses to Virus-Like Particle Vaccines

Abstract

Induction of persistent antibody responses by vaccination is generally thought to depend on efficient help by T follicular helper cells. Since the T helper cell response to HIV Env may not be optimal, we explored the possibility of improving the HIV Env antibody response to virus-like particle (VLP) vaccines by recruiting T helper cells induced by commonly used licensed vaccines to provide help for Env-specific B cells. B cells specific for the surface protein of a VLP can internalize the entire VLP and thus present peptides derived from the surface and core proteins on their major histocompatibility complex class II (MHC-II) molecules. This allows T helper cells specific for the core protein to provide intrastructural help for B cells recognizing the surface protein. Consistently, priming mice with an adjuvanted Gag protein vaccine enhanced the HIV Env antibody response to subsequent booster immunizations with HIV VLPs. To harness T helper cells induced by the licensed Tetanolpur vaccines, HIV VLPs that contained T helper cell epitopes of tetanus toxoid were generated. Tetanol-immunized mice raised stronger antibody responses to immunizations with VLPs containing tetanus toxoid T helper cell epitopes but not to VLPs lacking these epitopes. Depending on the priming immunization, the IgG subtype response to HIV Env after the VLP immunization could also be modified. Thus, harnessing T helper cells induced by other vaccines appears to be a promising approach to improve the HIV Env antibody response to VLP vaccines.IMPORTANCE Induction of HIV Env antibodies at sufficient levels with optimal Fc effector functions for durable protection remains a challenge. Efficient T cell help may be essential to induce such a desirable antibody response. Here, we provide proof of concept that T helper cells induced by a licensed vaccine can be harnessed to provide help for HIV Env-specific B cells and to modulate the Env-specific IgG subtype response.

Keywords: Env; HIV; IgG subtype; T helper cells; VLP; antibody; intrastructural help; vaccine.

FIG 1

Mechanism of intrastructural help. Env-specific B cells encountering HIV virions or VLPs take up the entire particle in a BCR-dependent manner and present Env- and Gag-derived T helper cell epitopes on their MHC-II molecules. Therefore, Gag-specific T helper cells can provide help to Env-specific B cells.

FIG 2

Intrastructural help by Gag immunogens. (A) BALB/c mice (n = 4 per group) were immunized once with 1 μg Gag and the indicated dose (μg) of poly(ICLC) (pICLC) or with 25 μg Gag DNA vaccine by i.m. DNA electroporation. Three weeks later, the antibody responses against Gag were analyzed at 1:500 serum dilutions. Shown are the mean values with the SEM for logarithmically transformed values for Gag IgG1 and IgG2a. *, P < 0.05; **, P < 0.01; ****, P < 0.001 (vaccine groups versus nonprimed; one-way ANOVA with Tukey's posttest). (B) Gag-specific CD4⁺ T cell responses were analyzed by intracellular cytokine staining for the indicated cytokines 2 weeks after a single i.m. injection of 1 μg Gag, 10 μg poly(ICLC) (pICLC), or the combination of Gag and pICLC or a single i.m. electroporation of 25 μg of a Gag DNA vaccine into BALB/c mice. Shown are the mean values with SEM for four animals per group (⁺, P < 0.05 versus PBS, Gag, and pICLC for IFN-γ; *, P < 0.05 versus PBS for IL-2; ##, P < 0.001 versus PBS and pICLC for TNF-α; #, P < 0.05 versus Gag for TNF-α [one-way ANOVA with Tukey's posttest]). (C) BALB/c mice (n = 11 or 12 per group) were immunized at weeks 0 and 4 with 1 μg Gag or 10 μg pICLC alone or in combination or with the Gag DNA vaccine. All primed and nonprimed mice were boosted at weeks 8 and 12 with the same VLP preparation containing Env and Gag, and naive sera were taken 1 week before the first immunization. (D) Antibody responses to Gag at 3 weeks after the second priming immunization at a serum dilution of 1:1,000. Shown are the mean values with SEM for 11 or 12 animals from two independent experiments. The dashed line represents the background of naive sera for Gag antibodies. For IgG1, *, P < 0.05 versus nonprimed; ⁺⁺⁺⁺, P < 0.001 versus nonprimed, Gag, pICLC, and Gag DNA (one-way ANOVA with Tukey's posttest). For IgG2a, ****, P < 0.0001 versus nonprimed, Gag, and pICLC; ⁺⁺⁺, P < 0.001 versus Gag DNA (one-way ANOVA with Tukey's posttest). (E) Env-specific antibody responses 2 weeks after the second VLP booster immunization. Shown are the mean values with SEM for logarithmically transformed HIV Env antibody concentrations in 11 or 12 animals from two independent experiments. For IgG1, *, P < 0.05 versus pICLC (Kruskal-Wallis test with Dunn's posttest). For IgG2a, **, P < 0.01 versus nonprimed, Gag, and pICLC; ⁺⁺⁺, P < 0.001 versus Gag and pICLC (Kruskal-Wallis test with Dunn's posttest). The dashed lower and upper lines represent the detection limits of the HIV Env-specific IgG2a and IgG1 antibody levels, respectively. (F) Env-specific IgG2a/IgG1 ratios 2 weeks after the second VLP immunization. The bars represent the median of the ratios for all animals of each group that were positive for both Env-specific IgG1 and IgG2a antibodies (open symbols). Samples that were under the limit of detection for IgG2a and/or IgG1 are shown by closed symbols. *, P < 0.05; **, P < 0.01 (Kruskal-Wallis test with Dunn's posttest; vaccine groups versus nonprimed).

FIG 3

Intrastructural help by tetanus toxoid-specific T helper cells. (A) Map of Gag expression plasmids encoding wild-type Gag, an FrC-Gag fusion protein, or an p2D1-Gag fusion protein. (B) Western blot analyses of wild-type VLPs (wt-VLP) and FrC-VLPs with antibodies detecting the indicated proteins. X, positions of lanes removed from the Western blots. (C to F) Env-specific antibody responses in BALB/c mice primed twice with Tetanol and boosted once or twice with FrC-VLPs (C), BL/6 mice primed twice with Tetanol and boosted once with FrC-VLPs (D), BALB/c mice primed twice with Tetanol and boosted once or twice with wt-VLPs (E), and BL/6 mice primed twice with Tetanol and boosted once with wt-VLPs (F). The dashed lines represent the background observed in naive animals. Shown are the means and the SEM of logarithmically transformed RLU for 4 animals per group at a 1:250 serum dilution (C and E) and for 6 animals per group at a 1:100 serum dilution (D and F). ***, P < 0.001 (unpaired t test).

FIG 4

Enhancement of the proliferative response of naive antigen-specific B cells by intrastructural help. (A) BL/6 mice were primed s.c. twice over a 4-week interval with 10 μl of Tetanol per animal. Ten days after the last Tetanol injection, CD4 cells were depleted. Two weeks after the last Tetanol injection, naive B cells from SW-HEL mice were labeled with CFSE and adoptively transferred into nonprimed or Tetanol-primed mice with or without CD4 cell depletion. The acceptor animals were injected i.v. with FrC-VLPs containing membrane-anchored HEL (HEL-VLP). Three days later, splenocytes of these mice and of untreated controls were stained with Alexa 647-conjugated HEL and anti-B220–APC-eFluor780 antibody. (B to F) Gating strategy for the analysis of in vivo proliferating CFSE⁺ HEL⁺ B220⁺ B cells. (B) Gating on lymphoid cells according to their forward scatter and sideward scatter. (C) Elimination of doublets by FSC-A versus. FSC-H gating. (D) A dump channel was used to exclude autofluorescent cells among B220-positive cell singlets. (E) HEL- and CFSE-positive cells were selected. (F) Gating on proliferating CFSE⁺ HEL⁺ B220⁺ B cells. Representative data are shown. (G) The bars represent mean values of percent proliferating cells for each experimental group; values for individual mice are also shown. ***, P < 0.001 (one-way ANOVA with Tukey's posttest). (H) The bars represent expansion of HEL-specific B cells (percentage of HEL⁺ B220⁺ B cells among the total B220⁺ B cell population) in nonprimed and Tetanol-primed animals versus controls. **, P < 0.001; ***, P < 0.001 (one-way ANOVA with Tukey's posttest; vaccine groups versus control).

FIG 5

Characterization of p2D1-VLP. (A) Wild-type VLPs (wt-VLP), FrC-VLPs, and p2D1-VLPs were analyzed by Western blotting with an antiserum against gp120 of HIV Env (α Env), a monoclonal antibody against p24 of Gag (α Gag), and a monoclonal antibody against FrC and p2D1 (α TT). (B to D) p2D1-VLPs containing or lacking Env (p2D1-VLPΔEnv) were immunoprecipitated with 2G12 or mock treated (Mock IP). Precipitates were analyzed by Western blotting with the antibodies described for panel A. Equal amounts of p2D1-VLPs that were used for the immunoprecipitation were analyzed directly by SDS-PAGE (Input). The additional bands at 25 kDa in immunoprecipitation samples are due to protein G that was coeluted from the beads.

FIG 6

Requirements for the modulation of the Env-specific IgG subtype response by intrastructural help. (A) BALB/c mice were immunized at weeks 0 and 4 with Tetanol or DNA vaccines expressing tetanus toxoid or HIV Gag. All primed and nonprimed mice were boosted at weeks 8 and 12 with p2D1-VLPs or wild-type VLPs (wt-VLP). (B) Total Env-specific antibody concentrations in sera of mice primed with the indicated immunogens 2 weeks after the second immunization with the indicated VLPs. Shown are the mean values with SEM of logarithmically transformed HIV Env antibody concentrations in the sera of 12 to 18 animals per group from three independent experiments. *, P < 0.05; **, P < 0.01, ***, P < 0.001; ****, P < 0.0001 (Kruskal-Wallis test with Dunn's posttest; vaccine groups versus nonprimed). (C to F) Env specific IgG1 (C), IgG2a (D), IgG2b (E), and IgG3 (F) subtypes at 2 weeks after the second VLP immunization in animals boosted with p2D1-VLPs versus those boosted with wt-VLPs. Shown are the mean values with SEM of logarithmically transformed concentrations of Env-specific IgG subtypes in sera of 12 to 18 animals per group from three independent experiments. *, P < 0.05; **, P < 0.01, ***, P < 0.001; ****, P < 0.0001 (Kruskal-Wallis test with Dunn's posttest; vaccine groups versus nonprimed). (B, D, and E) The dashed lines represent the background values for naive sera. (C and F) The background for naive sera lies on x axis at a value of 1.

FIG 7

Characterization of tetanus toxoid-specific cellular immune responses. BALB/c mice were immunized i.m. once with PBS, Tetanol, or TT DNA. (A) TT-specific T cell responses 2 weeks after immunization as determined by an IFN-γ ELISPOT assay with splenocytes stimulated with MHC-II-restricted p2 and p30 T helper peptides derived from TT. Shown are the mean values with SEM of spot-forming cells (SFC) producing IFN-γ per 200,000 splenocytes for 8 animals per group out of two independent experiments. **, P < 0.01 (one-way ANOVA with Tukey's posttest; vaccine groups versus PBS). (B) Percentage of splenic CD4⁺ T cells cell producing IFN-γ after in vitro stimulation with p2 and p30 T helper peptides as measured by intracellular cytokine staining. Mice (6 per group) were sacrificed 2 weeks after immunization. The cells were stained for surface expression of CD4 and for intracellular expression of IFN-γ. The background values for unstimulated cultures were subtracted. Shown are the mean values with SEM. *, P < 0.05 (one-way ANOVA with Tukey's posttest; vaccine groups versus PBS). (C and D) Tetanus-specific IL-4 (C) and IL-5 (D) cytokine secretion as determined by ELISA. Two weeks after immunization, splenocytes were stimulated with p2 and p30 T helper peptides for 48 h. The mean values with SEM for 10 animals out of two independent experiments are shown. *, P < 0.05 (Tetanol versus PBS); **, P < 0.01 (Tetanol versus TT DNA).

FIG 8

Fcγ receptor activation assays and durability of the humoral immune response. BALB/c mice (6 per group) were immunized twice at weeks 0 and 4 with Tetanol. Primed and nonprimed mice were boosted at weeks 8 and 12 with p2D1-VLPs. (A to C) FcγR activation by Env-specific Abs from sera taken 2 weeks after the second VLP immunization using FcγRII (A), FcγRIII (B), or FcγRIV (C) reporter cell lines secreting IL-2. Background values from parental P815 cells not expressing Env were subtracted from the values generated with P815Env cells to obtain the Env-specific Fcγ receptor activation. Shown are the mean IL-2 concentrations with SEM for 6 immune sera per group at a 1:100 dilution. *, P < 0.05 (Tetanol-primed versus nonprimed; unpaired t test). (D) Env IgG1 antibody levels were monitored for 10 weeks after the second VLP immunization. Shown are the mean values with SEM of logarithmically transformed RLUs in the sera of 6 animals per group (*, P < 0.05; **, P < 0.01, Tetanol-primed versus nonprimed, one-way ANOVA with Tukey's posttest).