Sublingual vaccination with influenza virus protects mice against lethal viral infection

Abstract

We assessed whether the sublingual (s.l.) route would be an effective means of delivering vaccines against influenza virus in mice by using either formalin-inactivated or live influenza A/PR/8 virus (H1N1). Sublingual administration of inactivated influenza virus given on two occasions induced both systemic and mucosal antibody responses and conferred protection against a lethal intranasal (i.n.) challenge with influenza virus. Coadministration of a mucosal adjuvant (mCTA-LTB) enhanced these responses and resulted in complete protection against respiratory viral challenge. In addition, s.l. administration of formalin-inactivated A/PR/8 plus mCTA-LTB induced systemic expansion of IFN-gamma-secreting T cells and virus-specific cytotoxic T lymphocyte responses. Importantly, a single s.l. administration of live A/PR/8 virus was not pathogenic and induced protection mediated by both acquired and innate immunity. Moreover, s.l. administration of live A/PR/8 virus conferred heterosubtypic protection against respiratory challenge with H3N2 virus. Unlike the i.n. route, the A/PR/8 virus, whether live or inactivated, did not migrate to or replicate in the CNS after s.l. administration. Based on these promising findings, we propose that the s.l. mucosal route offers an attractive alternative to mucosal routes for administering influenza vaccines.

Fig. 1.

A/PR/8 virus-specific Ab responses induced in serum and mucosal secretions after s.l. administration of inactivated A/PR/8/34 with or without adjuvant. Mice were vaccinated s.l. with inactivated A/PR/8 (20 μg) and mCTA/LTB (5 μg) as adjuvant at days 0 and 14. At 1 wk after the second immunization, serum and mucosal secretions were collected and A/PR/8 virus-specific Ab levels assessed by ELISA. (A) Anti-A/PR/8 virus-specific IgG and IgA Ab responses measured in sera and other mucosal secretions. Ab production levels are shown as reciprocal log₂ titer. (B) Mononuclear cells from spleen, lung nasal passage (NP), submandibular gland (SMG), lung, and small and large intestines (SI, LI) were prepared from s.l. immunized mice given formalin-inactivated virus with or without adjuvant. ELISPOT was used to count A/PR/8 virus-specific ASCs. (C) Levels of anti-A/PR/8 virus-specific Ab responses in pIgR^−/− mice immunized s.l. with inactivated A/PR/8 and mCTA/LTB. **, P < 0.01; ***, P < 0.001 vs. PBS-vaccinated group. Each group had five to seven mice. Data are representative of three separate experiments.

Fig. 2.

Induction of the IFN-γ-secreting T cells and influenza virus-specific CTL responses in mice after s.l. vaccination. (A) Analysis of IFN-γ-producing CD4⁺ and CD8⁺ T cells in spleen after two s.l. vaccinations with formalin-inactivated A/PR/8 virus with/without mucosal adjuvant (mCTA/LTB). Splenocytes from each vaccinated group were cultured with inactivated A/PR/8 for 4 days and T cells were gated out by anti-CD3 mAb. Then, IFN-γ-producing T cells were stained by anti-CD8 mAb. (B) At 2 wks after the second s.l. vaccination with killed influenza virus, splenocytes were cocultured with A/PR/8 virus-infected autologous splenocytes for 5 days. *, P < 0.05; **, P < 0.01; ***, P < 0.001 vs. PBS group. E:T, effector-to-target.

Fig. 3.

Vaccination s.l. with inactivated A/PR/8/34 with adjuvant protects mice against lethal challenge with homologous virus. BALB/c mice were immunized s.l. or i.n. with inactivated A/PR/8 (20 μg) plus mCTA/LTB (5 μg) as adjuvant at days 0 and 14. At 2 wks after immunization, mice were challenged i.n. with 20 LD₅₀ of influenza A/PR/8 virus. (A) Body weights and survival were monitored daily after challenge. (B) At day 3 after challenge, virus titers in BAL fluid were measured by plaque assay. **, P < 0.01 vs. mice vaccinated with A/PR/8 virus with adjuvant by i.n. (C) Protective efficacy against influenza infection in s.l. immunized wild-type and pIgR^−/− mice. N.D., not detectable. Each group consisted of five to seven mice. Data are representative of three separate experiments.

Fig. 4.

The s.l. route is safe and effective for vaccination with live A/PR/8 virus. Mice were vaccinated once s.l. with 800 pfu (2 LD₅₀) of live A/PR/8 virus. Two weeks later, mice were challenged i.n. with 20 LD₅₀ of influenza A/PR/8 virus. Virus titers and survival were monitored daily. (A) Levels of anti-A/PR/8 virus-specific IgG and IgA Ab responses in serum and mucosal secretions were determined by ELISA 2 wks after s.l. vaccination with live A/PR/8 virus. *, P < 0.05; **, P < 0.01; ***, P < 0.001 vs. PBS-vaccinated group. (B) Virus titers were measured 3 days after i.n. challenge. Survival was monitored for 14 days. (C) Three days after s.l. administration of 800 pfu (2 LD₅₀) of live A/PR/8, wild-type mice or MyD88^−/− mice were challenged i.n. with 20 LD₅₀ of A/PR/8 virus. Survival after i.n. challenge: ***, P < 0.001 vs. mice vaccinated with 800 pfu (2 LD₅₀) of live influenza virus. Each group had five to seven mice. Data are representative of three separate experiments.

Fig. 5.

Live and inactivated A/PR/8 viruses are not transported into the CNS after s.l. administration. (A) BALB/c mice were immunized i.n. or s.l. with acridinium-labeled, inactivated A/PR/8 virus 24 h before killing. All tissues were collected, homogenized, and analyzed for acridinium activity. Data are representative of three separate experiments, all of which showed the same pattern. (B) BALB/c mice were infected i.n. or s.l. with 20 LD₅₀ of live A/PR/8 virus. Viral RNA was detected 24 h later by RT-PCR with M2-specific primers (see Materials and Methods). ***, P < 0.001 vs. s.l. vaccinated group. (C) At 24 h after infection, lungs of mice challenged i.n. or s.l. with 20 LD₅₀ of live A/PR/8 were removed to determine virus titers (Left) and histopathological changes after H&E staining (Right). Pathological score was determined by using a blinded test (□, damage to epithelial cell layers; ■, infiltration of inflammatory mononuclear cells; ▧, alveolitis). *, P < 0.1; ***, P < 0.001 vs. naïve mice or s.l. vaccinated group. Each group had five to seven mice. Data are representative of two separate experiments.