B cells promote resistance to heterosubtypic strains of influenza via multiple mechanisms

Abstract

Immunity to heterosubtypic strains of influenza is thought to be mediated primarily by memory T cells, which recognize epitopes in conserved proteins. However, the involvement of B cells in this process is controversial. We show in this study that influenza-specific memory T cells are insufficient to protect mice against a lethal challenge with a virulent strain of influenza in the absence of B cells. B cells contribute to protection in multiple ways. First, although non-neutralizing Abs by themselves do not provide any protection to challenge infection, they do reduce weight loss, lower viral titers, and promote recovery of mice challenged with a virulent heterosubtypic virus in the presence of memory T cells. Non-neutralizing Abs also facilitate the expansion of responding memory CD8 T cells. Furthermore, in cooperation with memory T cells, naive B cells also promote recovery from infection with a virulent heterosubtypic virus by generating new neutralizing Abs. These data demonstrate that B cells use multiple mechanisms to promote resistance to heterosubtypic strains of influenza and suggest that vaccines that elicit both memory T cells and Abs to conserved epitopes of influenza may be an effective defense against a wide range of influenza serotypes.

Fig. 1. Failure of heterosubtypic protection in B cell deficient mice

A. C57BL/6 and μMT mice were infected or not with 300 EIU of X31, allowed to recover for 4 weeks and then challenged with 5000 EIU of PR8. B. Survival was monitored for 3 weeks after challenge infection. C. Weight loss was measured for 3 weeks following challenge infection. D. Viral titers were measured in the lungs 6 days after challenge infection. There were 4−5 mice per group in each panel. This experiment was performed twice with similar results.

Fig. 2. Influenza-specific memory T cells are not sufficient to provide efficient protection against heterosubtypic infection

C57BL/6 and μMT mice were infected or not with 300 EIU of X31, allowed to recover for 4 weeks and then challenged with 1000 EIU of PR8. A. Survival was monitored for 3 weeks after challenge infection. B. Weight loss was measured for 3 weeks following challenge infection. C. Viral titers were measured in the lungs 6 days after challenge infection. D. NP- and PA-specific CD8 T cells were detected in the spleen by flow cytometry on day 6 after challenge infection. The numbers in the panels indicate the frequency of antigen-specific cells in the CD8 population (mean ± standard deviation). The plots are gated on live CD8 T cells. E. The numbers of splenic NP-specific CD8 T cells were calculated on day 6 after challenge infection. There were 5 mice per group in each panel. This experiment was performed 3 times with similar results.

Fig. 3. Naïve B cells provide partial protection against heterosubtypic challenge

A. C57BL/6 and μMT mice were infected or not with 300 EIU of X31 and allowed to recover for 4 weeks. Some groups received 5 × 10⁷ purified splenic B cells 1 day prior to challenge with 1000 EIU of PR8. B. Survival was monitored for 3 weeks after challenge infection. C. Weight loss was measured for 3 weeks following challenge infection. D. Viral titers were measured in the lungs 6 days after challenge infection. E. NP- and PA-specific CD8 T cells were detected in the spleen by flow cytometry on day 6 after challenge infection. The numbers in the panels indicate the frequency of antigen-specific cells in the CD8 population (mean ± standard deviation). The plots are gated on live CD8 T cells. F. The numbers of splenic NP-specific CD8 T cells were calculated on day 6 after challenge infection. There were 5 mice per group in each panel. This experiment was performed 3 times with similar results. G-H. The serum titers of X31-specific (G) and PR8-specific (H) IgG were determined by ELISA in all groups 6 days after challenge infection and 21 days after challenge infection.

Fig. 4. X31-immune serum provides significant protection against heterosubtypic challenge in X31-immune μMT mice

A. μMT mice were infected or not with 300 EIU of X31 and allowed to recover for 4 weeks. Mice then received 400 μl of serum from either normal donors or from X31-immune donors 1 day prior to challenge with 1000 EIU of PR8. B. Survival was monitored for 3 weeks after challenge infection. C. Weight loss was measured for 3 weeks following challenge infection. D. Viral titers were measured in the lungs 6 days after challenge infection. E. NP- and PA-specific CD8 T cells were detected in the spleen by flow cytometry on day 6 after challenge infection. The numbers in the panels indicate the frequency of antigen-specific cells in the CD8 population (mean ± standard deviation). The plots are gated on live CD8 T cells. F. The numbers of splenic NP-specific CD8 T cells were calculated on day 6 after challenge infection. There were 5 mice per group in each panel. This experiment was performed 3 times with similar results.

Fig. 5. Naïve B cells, immune serum and memory T cells act together to protect against heterosubtypic challenge

A. C57BL/6 and μMT mice were infected or not with 300 EIU of X31 and allowed to recover for 4 weeks. Mice then received 5 × 10⁷ splenic B cells alone or with 400 μl of serum from X31-immune donors 1 day prior to challenge with 1000 EIU of PR8. B. Survival was monitored for 3 weeks after challenge infection. C. Weight loss was measured for 3 weeks following challenge infection. D. Viral titers were measured in the lungs 6 days after challenge infection. E. NP-specific CD8 T cells were detected in the spleen by flow cytometry on day 6 after challenge infection. The numbers in the panels indicate the frequency of NP-specific cells in the CD8 population (mean ± standard deviation). The plots are gated on live CD8 T cells. F. The numbers of splenic NP-specific CD8 T cells were calculated on day 6 after challenge infection. There were 5 mice per group in each panel.

Fig. 6. Antibodies against X31 cross-react with internal proteins of PR8

A-B. Antibodies in serum from X31-immune mice (A) or PR8-immune mice (B) were allowed to bind ELISA plates that were coated with proteins from disrupted X31 or PR8 viruses. Antibodies in naïve serum, X31-immune serum or serum from mice immunized with recombinant M2e-NP fusion protein were allowed to bind plates coated with purified M2e peptide (C) or purified recombinant NP protein (D). All plates were developed with anti-mouse IgG.