GM-CSF increases mucosal and systemic immunogenicity of an H1N1 influenza DNA vaccine administered into the epidermis of non-human primates

Abstract

Background: The recent H5N1 avian and H1N1 swine-origin influenza virus outbreaks reaffirm that the threat of a world-wide influenza pandemic is both real and ever-present. Vaccination is still considered the best strategy for protection against influenza virus infection but a significant challenge is to identify new vaccine approaches that offer accelerated production, broader protection against drifted and shifted strains, and the capacity to elicit anti-viral immune responses in the respiratory tract at the site of viral entry. As a safe alternative to live attenuated vaccines, the mucosal and systemic immunogenicity of an H1N1 influenza (A/New Caledonia/20/99) HA DNA vaccine administered by particle-mediated epidermal delivery (PMED or gene gun) was analyzed in rhesus macaques.

Methodology/principal findings: Macaques were immunized at weeks 0, 8, and 16 using a disposable single-shot particle-mediated delivery device designed for clinical use that delivers plasmid DNA directly into cells of the epidermis. Significant levels of hemagglutination inhibiting (HI) antibodies and cytokine-secreting HA-specific T cells were observed in the periphery of macaques following 1-3 doses of the PMED HA DNA vaccine. In addition, HA DNA vaccination induced detectable levels of HA-specific mucosal antibodies and T cells in the lung and gut-associated lymphoid tissues of vaccinated macaques. Importantly, co-delivery of a DNA encoding the rhesus macaque GM-CSF gene was found to significantly enhance both the systemic and mucosal immunogenicity of the HA DNA vaccine.

Conclusions/significance: These results provide strong support for the development of a particle-mediated epidermal DNA vaccine for protection against respiratory pathogens such as influenza and demonstrate, for the first time, the ability of skin-delivered GM-CSF to serve as an effective mucosal adjuvant for vaccine induction of immune responses in the gut and respiratory tract.

Figure 1. Particle-mediated DNA vaccine delivery into the epidermis of rhesus macaques using the disposable commercial prototype ND10 device.

Plasmid DNA encoding (A) the gene for influenza A/New Caledonia/20/99 haemagglutinin (HA) was administered alone or in combination with (B) a plasmid encoding the gene for rhesus macaque granulocyte-macrophage colony stimulating factor (rhGM-CSF) co-formulated onto 1–3 µM gold particles. (C, D) The ND10 delivery device consists of a cassette containing 2.0 µg plasmid (1.8µg HA DNA+0.2µg GM-CSF DNA) coated onto 1.0 mg of gold particles, a safety catch that is released when the device is held firmly against the skin surface, and an actuation button that breaks the tip off a gas microcylinder and releases helium at high pressure. Release of the helium ruptures the cassette membrane, entrains the DNA-coated gold particles into the helium jet, and propels them directly into cells in the skin. (E) Vaccinations were targeted to the skin located on the upper inner thigh adjacent to the inguinal lymph node. Immediately following vaccination, vaccination sites are easily visualized as red (erythema) targets in the skin. (F) The erythema is transient (24 hours) and vaccination sites faded but were still discernible at 8 weeks post-vaccination. (G) Shown is gold particle penetration into the epidermal and dermal skin layers in a representative histological cross-section of a skin biopsy collected 10 minutes after ND10 delivery.

Figure 2. Induction of virus neutralizing antibodies in the sera of macaques following immunization with either the unadjuvanted, or rhGM-CSF adjuvanted, PMED HA DNA vaccine.

Influenza A/New Caledonia/20/99 virus-neutralizing antibodies present in serum of macaques at the indicated times following vaccination with HA DNA alone (black bars) or HA DNA co-delivered with rhGM-CSF (hatched bars) were measured using a standard hemagglutination inhibition (HI) assay. (A) Serum HI titers measured in individual animals at the indicated times following the primary immunization at week 0. Bars indicate the frequency of animals in each group (n = 6) exhibiting a HI titer greater than or equal to 1/20. Numbers above each bar indicate the actual number of macaques in each group used to calculate the frequencies. (B) Serum HI titers were measured in vaccinated animals following boosting doses of vaccine at weeks 8 and 16 (indicated by the arrows). Bars indicate the means (± SEM) calculated for each group of at the indicated times following vaccination. Indicated P values were determined by the Mann-Whitney U test (two-tailed).

Figure 3. Co-delivery of the GM-CSF genetic adjuvant enhances HA-specific T cell responses in the peripheral blood of PMED HA DNA vaccinated animals.

PBMCs were isolated from the blood of macaques at the indicated times following the initial vaccination with HA DNA (black bars) or HA DNA+rhGM-CSF (hatched bars). (A) T cell proliferative responses in vaccinated macaques were measured by incubating PBMCs with recombinant A/New Caledonia/20/99 HA protein (1 µg/well) for 6 days prior to the addition of 1 µCi tritiated thymidine. Bars represent the mean (± SEM) stimulation index calculated for each group of animals. (B) The number of HA-specific T cells present in the blood of macaques after HA DNA vaccine priming, the first boost, and second boost was measured by IFN-γ ELISPOT assay. Indicated P values were determined using the Mann-Whitney U test (two-tailed).

Figure 4. Induction of mucosal antibody responses in the respiratory tract of PMED HA DNA vaccinated macaques.

Bronchoalveolar lavage fluid (BALF) and tracheal swab samples were collected from macaques prior to immunization and again at 4 weeks following each boosting dose of the HA DNA vaccine, ± rhGM-CSF. IgG antibody responses in BALF (A) and IgA antibody responses in tracheal swabs (B) against the A/New Caledonia/20/99 hemagglutinin protein were detected by ELISA. Data are reported as the O.D. measured at 450 nm for each sample, diluted 1∶20 in PBS, from individual immunized animals.

Figure 5. The GM-CSF genetic adjuvant increases both the magnitude and breadth of mucosal T cell responses elicited in the lungs and guts following PMED HA DNA vaccination.

HA-specific T cell responses in the (A) lung and (B) gut mucosa of macaques were determined by IFN-γ ELISpot assay 4–11 weeks after the final vaccination with HA DNA (solid bars) or HA DNA+GM-CSF (hatched bars). Individual bars represent the peak number of HA-specific IFN-γ T cells detected in the jejunum and lung tissue of individual animals 4–11 weeks after the final vaccination. (C) Breadth of the IFN-γ T cell response in PBMC, lung, and gut. HA-specific T cell responses in the indicated tissues were measured using a standard IFN-γ ELISpot assay with 6 individual pools of overlapping peptides (11 amino acid overlaps, 103 15-mers per pool) comprising the entire amino acid sequence of the influenza A/New Caledonia/20/99 HA protein. The percent contribution of each peptide-pool specific response to the total response was determined by dividing the mean number of IFN-γ spot forming cells (SFC) measured against each individual peptide pool by the sum of the response against all peptide pools. Results represent the average of 2 time-points tested after the 3^rd DNA dose (weeks 19 and 23). *Measurement below positive threshold level for the assay.

Figure 6. Generation of multifunctional T cells in the peripheral blood and lungs of macaques following PMED HA DNA vaccination.

Intracellular cytokine staining was performed on BALF collected from both vaccine groups at 4 weeks after the first boost (panels on left), as well as PBMCs from both vaccine groups collected 4 weeks after the second boost (panels on right) to measure the expression of IFN-γ, TNF-α, and IL-2 by CD4+ and CD8+ T cells after stimulation with overlapping peptides derived from the A/New Caledonia/20/99 HA protein. Bar charts show the mean total percentage (+/− SEM) of CD4+ and CD8+ T cells in the lungs (A) and blood (B) of animals from the unadjuvanted (black bars) and GM-CSF adjuvanted (hatched bars) found to express IFN-γ, TNF-α, or IL-2 following HA peptide stimulation. Indicated P values were determined using the Mann-Whitney U test (two-tailed). Stacked bar charts show the mean proportion of cells producing IFN-γ (black), TNF-α (gray), or IL-2 (hatched) to the total HA-specific CD4+ and CD8+ T cell response detected in the lung (C, E) and blood (D, F) of animals from the unadjuvanted and GM-CSF adjuvanted vaccine groups. Pie charts show the proportion of HA-specific CD4+ and CD8+ T cells in the lung (G) and blood (H) from both vaccine groups positive for the different combinations of one, two, or three cytokines.