Neonatal Neutrophil-mediated Control of Bordetella pertussis Is Disrupted by Pertussis Toxin

Abstract

The increased susceptibility of infants and young children to some diseases has often been explained as the neonatal immune system (NIS) being incomplete and/or underdeveloped. However, our recent work demonstrated that neonatal mice could clear a Bordetella pertussis (Bp) strain lacking pertussis toxin (PTx) (BpΔptx) much more efficiently than adult mice, indicating that the NIS can be extremely effective, but this ability is highly sensitive to being blocked by PTx. In this article, we investigated immunological mechanisms by which neonates efficiently and rapidly clear BpΔptx to better understand how the NIS functions and how PTx disrupts it. Depleting neutrophils, or blocking their recruitment, inhibited pups' ability to rapidly clear BpΔptx, revealing a critical role for neutrophils. Pups deficient in complement (C3-/-) failed to recruit neutrophils and did not efficiently clear BpΔptx but recovered these abilities upon treatment with C3a. Neutrophil depletion in C3-/- pups led to further failure to control BpΔptx, suggesting that neutrophils and complement have independent roles in rapid clearance of BpΔptx. Depleting or disrupting neutrophils and complement had negligible effect on the rapid growth of wild-type Bp, indicating that PTx blocks these otherwise highly effective aspects of the NIS.

Graphical abstract

FIGURE 1.

Neutrophils are required for rapid control of BpΔptx in neonatal lungs. (A) Schematic of anti-Ly6G treatment and bacterial inoculation. Created using Biorender.com. (B–D) Neutrophil counts (CD45⁺Ly6G⁺CD11b⁺) per lungs and representative density plots of neutrophils (display percent neutrophils from total CD45⁺ populations) from lungs of pups treated with PBS or anti-Ly6G at 2 h (B), 1 dpi (C), and 3 dpi (D). Black, PBS challenged; blue, WT Bp challenged; red, BpΔptx challenged; solid, PBS injected; striped, anti-Ly6G injected. Statistical significance was calculated via unpaired Student t test. *p < 0.0332, **p < 0.0021, ****p < 0.0001. (B–D) Experiments were completed in triplicate and pooled. Error bars show SEM (n = 5–6 pups per treatment per timepoint). CFU were recovered from lungs at 2 h, 1 dpi, and 3 dpi. (E) Blue line, WT Bp challenged; red line, BpΔptx challenged; solid line, PBS injected; dashed line, anti-Ly6G injected. The experiments were completed in triplicate and pooled. Error bars show SEM (n = 5–6 pups per treatment per timepoint). Statistical significance was calculated via two-way ANOVA. *p < 0.05, ****p < 0.0001.

FIGURE 2.

Neutrophils are critical to production/regulation of neonatal inflammatory cytokines. (A–C) Analysis of inflammatory cytokines from lung supernatants at 2 h (A), 1 dpi (B), and 3 dpi (C) (n = 3–4 pups per treatment per timepoint).

FIGURE 3.

CXCL1 mediates neonatal neutrophil recruitment to the lungs. (A–C) Neutrophils (CD45⁺Ly6G⁺CD11b⁺) per lung at 2 h (A), 1 dpi (B), and 3 dpi (C). Top panels display total neutrophil numbers in all groups. Middle panels are representative histograms of pups challenged with WT Bp and treated with PBS (black) or anti-CXCL1 (blue) at each timepoint. Bottom panels are representative histogram of pups challenged with BpΔptx and treated with PBS (black) or anti-CXCL1 (red) at each timepoint. (D) CFU recovered from lungs at 2 h, 1 dpi, and 3 dpi. Blue line, WT Bp challenged; red line, BpΔptx challenged; solid line, PBS injected; dashed line, anti-CXCL1 injected. The experiments were completed in triplicate and pooled. Error bars show SEM (n = 5–6 pups per treatment per timepoint). Statistical significance was calculated via one-way ANOVA. *p < 0.05, **p < 0.0021, ***p < 0.0002, ****p < 0.0001.

FIGURE 4.

C3 contributes to neutrophil recruitment to neonatal lungs. (A and B) CFU per lung from C57BL/6 (solid line) or C3^−/− pups (dashed line) challenged with WT Bp (blue) (A) or BpΔptx (red) (B) at 2 h, 1 dpi, and 3 dpi. (C and D) Neutrophils recovered per lung from C57BL/6 (solid line) or C3^−/− pups (dashed line) challenged with WT Bp (blue) (C) or BpΔptx (red) (D) at 2 h, 1 dpi, or 3 dpi. Neutrophils recovered from lungs of C57BL/6 (solid red bar) or C3^−/− pups challenged with BpΔptx (red and white striped bar) or BpΔptx with 6.4 ng (red and black striped bar) C3a at 1 and 3 dpi. (E) Dashed line represents average neutrophils per lung in C57BL/6 pups at indicated timepoint. (F) CFU per lung from C57BL/6 (solid red bar) or C3^−/− pups challenged with BpΔptx (red and white striped bar) or BpΔptx with 6.4 ng C3a (red and black striped bar) at 1 and 3 dpi. (G) Representative density plots of neutrophils (CD45⁺Ly6G⁺CD11b⁺ from C3^−/− pups challenged with BpΔptx or BpΔptx + C3a at 1 and 3 dpi. The experiments were completed in triplicate and pooled. Error bars show SEM (n = 5–6 pups per treatment per timepoint). Statistical significance was calculated via two-way ANOVA. *p < 0.05, **p < 0.0021, ***p < 0.0002, ****p < 0.0001.

FIGURE 5.

Neonatal neutrophils and complement component 3 have individual contributions to control of B. pertussis at early timepoints but are disrupted by PTx. (A) Inoculation of C57BL/6, PAD4^+/+, and PAD4^−/− P5 pups with WT Bp or BpΔptx and collected lungs at 3 dpi. Incubation of lung cells with pHrodo green–stained WT Bp (blue) and BpΔptx (red) does not show significant effects of PTx on neutrophil phagocytosis. (B) Percentage pHrodo+ signals of total neutrophils per sample. The control samples were incubated on ice, and the experimental group was incubated at 37°C. (C) Total events shown represent events from experimental group incubated at 37°C subtracted from control group incubated on ice. (D and E) CFU per lung from C57BL/6 or C3^−/− pups treated with PBS (solid) or anti-Ly6G (striped) and challenged with WT Bp (blue) or BpΔptx (red) at 1 dpi (D) and 3 dpi (E). The experiments were completed in triplicate and pooled. Error bars show SEM (n = 5–6 pups per treatment per timepoint). Statistical significance was calculated via two-way ANOVA. ns, p > 0.05, *p < 0.0332, ***p < 0.0002, ****p < 0.0001.