Cross-Protective Immune Responses Induced by Sequential Influenza Virus Infection and by Sequential Vaccination With Inactivated Influenza Vaccines

Abstract

Sequential infection with antigenically distinct influenza viruses induces cross-protective immune responses against heterologous virus strains in animal models. Here we investigated whether sequential immunization with antigenically distinct influenza vaccines can also provide cross-protection. To this end, we compared immune responses and protective potential against challenge with A(H1N1)pdm09 in mice infected sequentially with seasonal A(H1N1) virus followed by A(H3N2) virus or immunized sequentially with whole inactivated virus (WIV) or subunit (SU) vaccine derived from these viruses. Sequential infection provided solid cross-protection against A(H1N1)pdm09 infection while sequential vaccination with WIV, though not capable of preventing weight loss upon infection completely, protected the mice from reaching the humane endpoint. In contrast, sequential SU vaccination did not prevent rapid and extensive weight loss. Protection correlated with levels of cross-reactive but non-neutralizing antibodies of the IgG2a subclass, general increase of memory T cells and induction of influenza-specific CD4+ and CD8+ T cells. Adoptive serum transfer experiments revealed that despite lacking neutralizing activity, serum antibodies induced by sequential infection protected mice from weight loss and vigorous virus growth in the lungs upon A(H1N1)pdm09 virus challenge. Antibodies induced by WIV vaccination alleviated symptoms but could not control virus growth in the lung. Depletion of T cells prior to challenge revealed that CD8+ T cells, but not CD4+ T cells, contributed to cross-protection. These results imply that sequential immunization with WIV but not SU derived from antigenically distinct viruses could alleviate the severity of infection caused by a pandemic and may improve protection to unpredictable seasonal infection.

Keywords: antigenically distinct influenza virus strains; cross-protection; immune mechanism; non-neutralizing antibody; sequential vaccination.

Figure 1

Weight loss, survival rate, and lung virus titer of immunized mice after A(H1N1)pdm09 virus challenge. Naïve mice (n = 10) were sequentially infected with sublethal doses of two different strains (PR8 and then X31) of live virus (LV) with 28 days interval or were sequentially immunized with vaccines (WIV, SU) derived from these virus strains and then challenged with virus A/California/7/2009 (H1N1)pdm09. After challenge, mice (n = 5) were monitored daily for weight loss (A) and survival (B) for a period of 14 days. On day 3 post-challenge, lung virus titers in 5 mice/group were determined by titration on MDCK cells (C). ^*p < 0.05, Mann-Whitney U-test. The dashed line represents the limit of detection.

Figure 2

Cross-reactive antibody responses induced by sequential infection or immunization. On day 28 (the day of the second infection or immunization; A) and on day 56 (the day of sacrifice or challenge; B–F), serum samples and nasal washes were collected from the mice described in the legend to Figure 1. Anti-PR8 (HA/NA) IgG antibodies (A; n = 7) and Anti-X-31(HA/NA) IgG antibodies (B; n = 7) in serum samples were detected by ELISA using PR8 SU and X-31 SU for coating. Anti-H1N1pdm09-specific IgG (C; n = 15), IgG2a and IgG1 (D; n = 5) antibodies in serum samples were detected by ELISA. Microneutralization assay was used to determine the neutralizing ability of these antibodies toward A(H1N1)pdm09 virus (E; n = 5). Anti-H1N1pdm09 IgA antibody levels in nasal washes were determined by ELISA (F; n = 5). Data of individual animals (A–E) are depicted or mean values ± SEM (F) are given, ^**p < 0.01, ^***p < 0.001. Mann-Whitney U-test. The dashed line represents the limit of detection.

Figure 3

Memory T cell immune responses after sequential infection or immunization. Of the mice described in the legend to Figure 1, 5 animals/group were sacrificed 28 days after the second infection or immunization and spleen and lung were collected. (A) CD4⁺CD44⁺ and CD8⁺CD44⁺ memory T cells in spleen were determined by flow cytometry. (B) CD8⁺CD44⁺CD62L⁻ effector memory T cells (TEM) and CD8⁺CD44⁺CD62L⁺ central memory T cells (TCM) in spleen. Left: representative dot plots depicting CD44 and CD62L expression on spleen CD8 T cells. Right: percentages of spleen CD8 TEM and TCM + SEM. (n = 4 or 5 per group, representative of two experiments, Mann-Whitney U-test, ^*p < 0.05).

Figure 4

Influenza-specific T cell immune responses induced by sequential infection or immunization. (A) Splenocytes harvested on day 28 post the second infection/vaccination, were stimulated with A(H1N1)pdm09 WIV and anti-CD28 overnight in presence of protein transport inhibitor. Presence of intracellular IFNγ in CD4⁺CD44⁺ and CD8⁺CD44⁺ T cells was analyzed by flow cytometry. Left: representative dot plots of stimulated CD4 or CD8 T cells stained for CD44 and IFNγ. Right: percentages of IFNγ-producing cells among CD4⁺CD44⁺ and CD8⁺CD44⁺ T cells. (n = 4 or 5, representative of two experiments, Mann-Whitney U-test, ^*p < 0.05). (B) On day 28 post the second infection/immunization, NP_366−374 of PR8 virus was used to stimulate mouse splenocytes and IFNγ-producing CD8 T cells were enumerated by ELISPOT. (n = 5, Mann-Whitney U test, ^** p < 0.01). (C) A(H1N1)pdm09 NP_366−374-specific CD8 T cells in spleens of infected/immunized mice (n = 5) were determined by tetramer assay. Lymphocytes from the blood of mice (n = 2) infected with A(H1N1)pdm09 virus served as positive control.

Figure 5

The cross-protective potential of antibodies induced by sequential infection or immunization. Mice (n = 5) were primed with PR8 virus (10³TCID₅₀) or PR8 WIV (15 μg) and boosted with X-31 virus (10³TCID₅₀) or X-31 WIV (15 μg). Mice primed and boosted with PBS served as negative control and mice primed and boosted with A(H1N1)pdm09 WIV (15 μg) served as positive control. Sera from these mice were collected 4 weeks after boost, pooled and injected into naïve mice 1 day before challenge with A/California/7/2009 (H1N1)pdm09 virus. Body weight loss (A) was monitored daily for 6 days. Virus titers in the lung tissue (B) on day 6 post-challenge were determined by titration on MDCK cells. ^**p < 0.01, Mann-Whitney U-test. The dashed line represents limit of detection. NS, not significant.

Figure 6

The cross-protective potential of CD4 T cells and CD8 T cells induced by sequential infection or immunization. Mice were primed with PR8 virus (10³TCID₅₀) or PR8 WIV (15 μg) and then boosted with X-31 virus (10³TCID₅₀) or X-31 WIV (15 μg). Mice primed and boosted with PBS served as control. Anti-CD4 or anti-CD8 T cell depletion antibody or PBS were injected intraperitoneally into mice on day −1, 1, and 3 of A(H1N1)pdm09challenge. Weight loss (A) was monitored for 6 days and lung virus titers (B) were determined on day 6 post-infection by titration on MDCK cells. ^*p < 0.05, Mann-Whitney U test. The dashed line represents limit of detection.