Enhanced natural killer-cell and T-cell responses to influenza A virus during pregnancy

Abstract

Pregnant women experience increased morbidity and mortality after influenza infection, for reasons that are not understood. Although some data suggest that natural killer (NK)- and T-cell responses are suppressed during pregnancy, influenza-specific responses have not been previously evaluated. Thus, we analyzed the responses of women that were pregnant (n = 21) versus those that were not (n = 29) immediately before inactivated influenza vaccination (IIV), 7 d after vaccination, and 6 wk postpartum. Expression of CD107a (a marker of cytolysis) and production of IFN-γ and macrophage inflammatory protein (MIP) 1β were assessed by flow cytometry. Pregnant women had a significantly increased percentage of NK cells producing a MIP-1β response to pH1N1 virus compared with nonpregnant women pre-IIV [median, 6.66 vs. 0.90% (P = 0.0149)] and 7 d post-IIV [median, 11.23 vs. 2.81% (P = 0.004)], indicating a heightened chemokine response in pregnant women that was further enhanced by the vaccination. Pregnant women also exhibited significantly increased T-cell production of MIP-1β and polyfunctionality in NK and T cells to pH1N1 virus pre- and post-IIV. NK- and T-cell polyfunctionality was also enhanced in pregnant women in response to the H3N2 viral strain. In contrast, pregnant women had significantly reduced NK- and T-cell responses to phorbol 12-myristate 13-acetate and ionomycin. This type of stimulation led to the conclusion that NK- and T-cell responses during pregnancy are suppressed, but clearly this conclusion is not correct relative to the more biologically relevant assays described here. Robust cellular immune responses to influenza during pregnancy could drive pulmonary inflammation, explaining increased morbidity and mortality.

Fig. 1.

NK- and T-cell frequencies in pregnant and control women. (A) Gating strategy to define NK- and T-cell populations, using CD7 to define NK cells (51). FSC, forward scatter; SSC, side scatter; Avid, aqua live/dead viability dye. (B) NK, CD8⁺, and CD4⁺ cell frequencies in pregnant (P) and control (C) women at day 0 and day 7. Error bars reflect the median and interquartile range. (C) NK-cell, CD8⁺ T-cell, and CD4⁺ T-cell frequencies in pregnant women at day 0, day 7, and 6 wk postpartum (PP).

Fig. 2.

NK- and T-cell responses to pH1N1 virus and PMA/I in pregnant women and controls. (A) Representative plots showing NK-cell IFN-γ, MIP-1β, and CD107a expression. (B) NK-cell, CD8⁺ T-cell, and CD4⁺ T-cell IFN-γ, MIP-1β, and CD107a expression after pH1N1 virus stimulation in pregnant (P) and control (C) women. (C) NK-cell, CD8⁺ T-cell, and CD4⁺ T-cell IFN-γ, MIP-1β, and CD107a expression after PMA/I stimulation. Error bars reflect the median and interquartile range.

Fig. 3.

Polyfunctional NK- and T-cell responses to pH1N1 virus in pregnant women and controls. (A) NK-cell, CD8⁺ T-cell, and CD4⁺ T-cell responses in control (C) and pregnant (P) women after pH1N1 virus stimulation to compare the following: one function (IFN-γ, MIP-1β, or CD107a), any two functions, or all three functions at day 0 and day 7. Error bars reflect the median and interquartile range. (B) All polyfunctional combinations of either IFN-γ, MIP-1β, or CD107a. Statistical differences are represented by *P < 0.05 and **P < 0.005 for differences between pregnant and controls at day 0 and day 7. Orange indicates superior responses in pregnant women at day 0, and auburn indicates superior responses in pregnant women at day 7.

Fig. 4.

Longitudinal NK- and T-cell functional responses to pH1N1 virus and PMA/I in pregnant women. (A) NK-cell, CD8⁺ T-cell, and CD4⁺ T-cell IFN-γ, MIP-1β, and CD107a expression in pregnant women after pH1N1 virus stimulation at day 0, day 7, and 6 wk postpartum (PP). (B) IFN-γ, MIP-1β, and CD107a expression in pregnant women after PMA/I stimulation at day 0, day 7, and 6 wk postpartum (PP).