Mucosal immunization with integrase-defective lentiviral vectors protects against influenza virus challenge in mice

Abstract

Recent reports highlight the potential for integrase-defective lentiviral vectors (IDLV) to be developed as vaccines due to their ability to elicit cell-mediated and humoral immune responses after intramuscular administration. Differently from their integrase-competent counterpart, whose utility for vaccine development is limited by the potential for insertional mutagenesis, IDLV possess a mutation in their integrase gene that prevents genomic integration. Instead, they are maintained as episomal DNA circles that retain the ability to stably express functional proteins. Despite their favorable profile, it is unknown whether IDLV elicit immune responses after intranasal administration, a route that could be advantageous in the case of infection with a respiratory agent. Using influenza as a model, we constructed IDLV expressing the influenza virus nucleoprotein (IDLV-NP), and tested their ability to generate NP-specific immune responses and protect from challenge in vivo. We found that administration of IDLV-NP elicited NP-specific T cell and antibody responses in BALB/c mice. Importantly, IDLV-NP was protective against homologous and heterosubtypic influenza virus challenge only when given by the intranasal route. This is the first report demonstrating that IDLV can induce protective immunity after intranasal administration, and suggests that IDLV may represent a promising vaccine platform against infectious agents.

Figure 1. Schematic of lentiviral vectors.

(A) Lentiviral transfer plasmid that expresses either NP (pTY2-CMV-NP), GFP (pTY2-CMV-GFP) or no exogenous protein (pTY2-CMV-empty). The arrow indicates where a woodchuck hepatitis virus posttranscriptional regulatory element (WPRE) is added for the production of the pTY2-CMV-NP_WPRE and pTY2-CMV-GFP_WPRE constructs. (B) Lentiviral packaging plasmid, pCHelp/IN- and pCMVΔR8.2, which are differentiated by the presence of the mutation D116N that inactivates the function of the integrase protein in pCHelp/IN- (indicated by a star). (C) Envelope-expressing plasmid, pMD.G, which produces the vesicular stomatitis G protein (VSV-G). Abbreviations: cytomegalovirus promoter (CMV), long terminal repeat (LTR), deleted gag gene (ΔGag), packaging signal (Ψ), rev responsive element (RRE), central polypurine tract (cPPT), nucleoprotein (NP), green fluorescent protein (GFP), deleted unique 3′ region (ΔU3), bovine growth hormone polyadenylation signal (polyA), deleted packaging signal (ΔΨ), nonfunctional envelope (ΔEnv).

Figure 2. IDLV express NP protein in vitro.

(A) 293T cells transduced with IDLV-NP or ICLV-NP with a multiplicity of infection (MOI) of 3 were fixed and stained at day 2 post-transduction with a FITC-conjugated antibody against influenza A virus NP, then measured by flow cytometry. Untransduced cells were used as a negative control. The percentage of FITC-positive cells and mean fluorescence intensity (MFI) in each sample is indicated. (B) 293T cells were transduced for 1 day with IDLV-NP (+ or − WPRE) (MOI = 3), fixed and stained with a FITC-conjugated antibody against influenza A virus NP, and then measured by flow cytometry. The percentage of FITC-positive cells and MFI in each sample is indicated. ICLV-empty was used as a negative control (empty LV). (C) Lysates from cells transduced as described in panel B were collected on day 1, 2 or 3, and probed for influenza virus NP (∼56 kDa) by western blot. Equivalent amounts of protein (9 µg) from each sample were loaded. Lysate from cells infected with PR8 influenza virus (MOI = 0.5) was used as a positive control for protein expression.

Figure 3. IDLV generate dose-responsive and antigen-specific T cell responses in mice.

Groups of BALB/c mice (n = 3) were immunized i.m. with varying doses (TU/mouse) of IDLV-GFP, as indicated. Splenocytes were assayed on day 10 (A) or day 30 (B) post-immunization by ELISPOT for IFN-γ responses to an H-2k^d-restricted 9-mer GFP peptide, or to an unrelated peptide (not shown). Mice injected with PBS served as a negative control. The Kruskal-Wallis test was used to compare GFP levels between the dosing groups (overall p-value of 0.006). P-values from pairwise dose group comparisons based on the ANOVA test are shown.

Figure 4. Immunization with IDLV-NP induces NP-specific cell-mediated and humoral immune responses.

Separate groups of mice were inoculated either once (1X) or twice (2X), 4 weeks apart, with IDLV-NP via either the i.m. or i.n. route of administration. Mice that were injected i.m. with PBS served as a negative control. (A) Splenic T cell responses were evaluated 4 weeks following the final immunization by IFN-γ ELISPOT after restimulation with the immunodominant H-2k^d-restricted 9-mer NP peptide, or with an unrelated H-2k^d-restricted 9-mer peptide (not shown). (B) Serum was collected from each mouse prior to the first immunization and 4 weeks after the last immunization. NP-specific IgG in serially-diluted serum samples was measured by indirect ELISA assay and is expressed as endpoint titer. The lower limit of detection is indicated by the dotted line. For all panels, each dot represents one mouse, and bold lines indicate the mean.

Figure 5. Immunization with IDLV-NP induces NP-specific IgG responses that are increased upon boosting.

Mice were inoculated 2 times i.n. or i.m. with IDLV-NP, 4 weeks apart. Mice that received either PBS or IDLV-GFP 1 or 2 times i.n. served as a negative control. Serum was collected from each mouse prior to the first immunization, prior to the second immunization (1X), and 4 weeks after the second immunization (2X). NP-specific IgG in serially-diluted serum samples was measured by indirect ELISA assay and is expressed as endpoint titers, defined using the pre-immunization sera as a baseline. Each dot in the panel represents one mouse, and bold lines indicate the mean. The lower limit of detection is indicated by the dotted line. The Kruskal-Wallis test was used to compare antibody titers between the groups (overall p-value of <0.0001). P-values from pairwise group comparisons based on the ANOVA test are shown.

Figure 6. Immunization with IDLV protects mice from homologous influenza challenge.

Groups of mice (n = 4) were immunized 1 or 2 times, 4 weeks apart, with IDLV-NP via either the i.m. or i.n. route of administration. Mice that were injected i.m. with PBS served as a negative control, and mice that were infected i.n. with a sublethal dose of PR8 influenza virus served as a positive control for protection. Survival (A) and weight loss (B) were monitored for 11 days following i.n. challenge with a lethal dose (10 LD₅₀) of PR8 influenza virus 4 weeks after the final inoculation. Weight loss is presented as the average percentage of initial weight at the time of challenge. *indicates days when IDLV-NP 2X i.n. lost significantly less weight than any of the other immunization regimens (p<0.001, calculated using pairwise group comparisons, at each time point, based on ANOVA). Overall p-value for group comparisons was based on the Kruskal-Wallis test: p = 0.001 for day 3, p = 0.002 for day 4, p = 0.005 for day 5, p = 0.008 for day 6.

Figure 7. Immunization with IDLV protects mice from challenge with influenza virus in an antigen-specific manner.

Mice were inoculated 2 times i.n., 4 weeks apart, with IDLV-NP (n = 5) or IDLV-GFP (n = 4). Mice that received PBS 2 times i.n. (n = 4) served as a negative control for protection. Four weeks after the final administration, mice were challenged with a lethal dose (5 LD50) of the mouse-adapted A/Philippines/1982:PR8 influenza virus. Survival (A) and weight loss (B) were monitored for 10 days following challenge. Weight loss is presented as the average percentage of initial weight at the time of challenge. *indicates days when IDLV-NP 2X i.n. lost significantly less weight than any of the other immunization regimens (p<0.005, calculated using pairwise group comparisons, at each time point, based on the ANOVA test). Overall p-value for group comparisons was based on the Kruskal-Wallis test: p = 0.029 for day 6, p = 0.021 for day 7, p = 0.026 for day 8.

Figure 8. Immunization with IDLV-NP protects mice from challenge with a heterosubtypic influenza virus.

Groups of mice (n = 3) were inoculated with IDLV-NP 2 times i.n. or i.m., 4 weeks apart. Mice (n = 3) that received PBS 2 times i.n. served as a negative control, mice (n = 3) inoculated with 2 i.n. doses of IDLV-GFP, 4 weeks apart, served as a specificity control, and mice (n = 3) that were infected with a sublethal dose of the mouse-adapted A/Netherlands/602/2009 (pH1N1) influenza virus served as a positive control for protection. Six weeks after the final administration, mice were challenged with a lethal dose (>10 LD₅₀) of pH1N1 influenza virus and monitored for survival (A) and weight loss (B). P-values calculated using pairwise group comparisons, at each time point, were based on the ANOVA test. For weight comparisons, Day 3: p<0.05 IDLV-NP 2X i.n. versus PBS; p<0.01 IDLV-NP 2X i.n. versus IDLV-GFP 2X i.n. or IDLV-NP 2X i.m.; Day 4: p<0.05 IDLV-NP 2X i.n. versus PBS and IDLV-NP 2X i.m.; p<0.005 IDLV-NP 2X i.n. versus IDLV-GFP 2X i.n.; Day 5, p<0.05 IDLV-NP 2X i.n. versus IDLV-NP 2X i.m.; p = 0.005 IDLV-NP 2X i.n. versus IDLV-GFP 2X i.n. Overall p-value for group comparisons based on the Kruskal-Wallis test: p = 0.024 for day 3, p = 0.029 for day 4, p = 0.027 for day 5.