Bordetella pertussis infection exacerbates influenza virus infection through pertussis toxin-mediated suppression of innate immunity

Abstract

Pertussis (whooping cough) is frequently complicated by concomitant infections with respiratory viruses. Here we report the effect of Bordetella pertussis infection on subsequent influenza virus (PR8) infection in mouse models and the role of pertussis toxin (PT) in this effect. BALB/c mice infected with a wild-type strain of B. pertussis (WT) and subsequently (up to 14 days later) infected with PR8 had significantly increased pulmonary viral titers, lung pathology and mortality compared to mice similarly infected with a PT-deficient mutant strain (ΔPT) and PR8. Substitution of WT infection by intranasal treatment with purified active PT was sufficient to replicate the exacerbating effects on PR8 infection in BALB/c and C57/BL6 mice, but the effects of PT were lost when toxin was administered 24 h after virus inoculation. PT had no effect on virus titers in primary cultures of murine tracheal epithelial cells (mTECs) in vitro, suggesting the toxin targets an early immune response to increase viral titers in the mouse model. However, type I interferon responses were not affected by PT. Whole genome microarray analysis of gene expression in lung tissue from PT-treated and control PR8-infected mice at 12 and 36 h post-virus inoculation revealed that PT treatment suppressed numerous genes associated with communication between innate and adaptive immune responses. In mice depleted of alveolar macrophages, increase of pulmonary viral titers by PT treatment was lost. PT also suppressed levels of IL-1β, IL-12, IFN-γ, IL-6, KC, MCP-1 and TNF-α in the airways after PR8 infection. Furthermore PT treatment inhibited early recruitment of neutrophils and NK cells to the airways. Together these findings demonstrate that infection with B. pertussis through PT activity predisposes the host to exacerbated influenza infection by countering protective innate immune responses that control virus titers.

Figure 1. Influenza virus infection in B. pertussis-infected mice and effect of PT.

BALB/c mice were inoculated with WT or ΔPT B. pertussis (5×10⁵ CFU) or PBS and seven days later infected with influenza virus PR8 (600 PFU). (A) Mean pulmonary viral titers, (B) survival rate, and (C) weight changes were assessed for 10 days post virus inoculation. (D, E) Bacterial loads in virus-infected and control mice at the indicated time points, when virus was inoculated (D) 7 days after bacterial inoculation, or (E) concomitantly with bacteria. (F) Mean pulmonary viral titers (day 6 post virus inoculation) in mice inoculated with WT or ΔPT B. pertussis and infected with influenza PR8 (600 PFU) at the indicated times post B. pertussis inoculation. n = 3–4 mice/treatment group/time point (4–5 mice for the last time point in A–C). *Significantly different from control (P<0.05).

Figure 2. Effect of PT enzymatic activity on influenza infection in BALB/c and C57BL/6 mice.

Mice were intranasally treated with 100 ng purified PT or PT-9K/129G (PT*) or PBS and infected with influenza virus PR8 (600 PFU) 24 h later. Mean pulmonary viral titers, survival rate and weight changes in BALB/c (A–C) and C57BL/6 (D–F) mice assessed for 9–13 days post virus inoculation. n = 3 mice/treatment group/time point (4–5 mice for the last time point for survival assessment). *Significantly different from control (P<0.05).

Figure 3. Histopathological examination of lungs from PT-treated and control mice infected with influenza virus.

Lung sections of BALB/c mice inoculated with 100 ng of PT or with PBS and infected with influenza PR8 (600 PFU) 24 h later. Sections were prepared from (A–B) uninfected mice, or 3 and 6 days after virus inoculation from (C–D) PBS-treated or (E–F) PT-treated mice. Lungs were harvested and fixed immediately in 4% formalin. Sections were stained with hematoxylin-eosin and evaluated for density and location of cellular infiltrates. Sections include representative results from 3 mice/time point. Original magnification 10×. n = 3 mice/treatment group.

Figure 4. Effect of prior, concurrent and delayed PT treatment on influenza.

(A) C57BL/6 mice were intranasally treated with 100 ng of PT, PT* or with PBS and infected with influenza PR8 (600 PFU) at the indicated times post PT treatment. Mean pulmonary viral titers were determined 2 days post-virus inoculation. (B–D) Mean pulmonary viral titers of mice treated with 100 ng of PT or PBS (control) 24 h prior (B), concurrently with (C), or 24 h post (D) infection with influenza PR8. Titers were determined 2, 4 and 6 days post virus inoculation. n = 3 mice/treatment group. *Significantly different from control (P<0.05).

Figure 5. Effect of PT on influenza WSN virus infection in BALB/c mice and on mTECs.

(A) Groups of BALB/c mice (n = 4) were pretreated with 100 ng of PT or an equal volume of PBS and 24 h later inoculated with 500 PFU of influenza WSN virus. Mean pulmonary viral titers were determined 3 days post virus inoculation. (B) mTECs (3 wells per group) were treated with 1 nM or 5 nM PT or 5 nM PT* for 24 h, or left untreated. The cells were then washed and infected with WSN virus at an MOI of approximately 0.001. After 1 h incubation, the inoculum was removed, and the cells were washed with medium and incubated at 37°C. At the indicated times, the infected-cell supernatant was sampled, and infectious virus titers were determined by TCID50.

Figure 6. PT targets AM to enhance early viral titers and suppresses early pro-inflammatory airway responses.

Groups of BALB/c mice were intranasally inoculated with 100 µL clodronate liposomes (CL) to deplete AM or with PBS liposomes (PL) as a control. (A) Percent of resident AM and neutrophils in the airways (BAL) two days after CL and PL treatment. (B) CL- and PL-treated mice were intranasally treated 2 days later with 100 ng of PT or an equal volume of PBS and infected with influenza PR8 (600 PFU) virus 24 h after PT/PBS treatment. Mean pulmonary viral titers were determined on days 2 and 4 post virus inoculation. n = 4 mice/treatment group. *Significantly different from control (P<0.05).

Figure 7. PT suppresses early pro-inflammatory responses in the airways.

Groups of BALB/c mice were intranasally inoculated with 100 ng of PT or an equal volume of PBS as a control and infected with influenza PR8 (600 PFU) 24 h later. Cytokine concentrations and cellular recruitment was assessed two days after virus inoculation. (A) Lung (BAL) inflammatory cytokine levels assayed by multiplex bead array. (B) Numbers of lung immune cell populations in BAL fluid of PT-treated (PT/PR8) or PBS control-treated (PR8) mice infected with influenza virus (and in untreated mice). Neutrophils (CD11b⁺, Gr1^hi), alveolar macrophages (CD11b⁻, CD11c⁺, F4/80⁺), and NK cells (NK1.1⁺). n = 4 mice/treatment group. Significant differences are indicated by P value.