Earliest innate immune responses require macrophage RelA during pneumococcal pneumonia

Abstract

NF-κB regulates cytokine expression to initiate and control the innate immune response to lung infections. The NF-κB protein RelA is critical for pulmonary host defense during Streptococcus pneumoniae pneumonia, but the cell-specific roles of this transcription factor remain to be determined. We hypothesized that RelA in alveolar macrophages contributes to cytokine expression and host defense during pneumococcal pneumonia. To test this hypothesis, we compared mice lacking RelA exclusively in myeloid cells (RelA(Δ/Δ)) with littermate controls (RelA(F/F)). Alveolar macrophages from RelA(Δ/Δ) mice expressed no full-length RelA, demonstrating effective targeting. Alveolar macrophages from RelA(Δ/Δ) mice exhibited reduced, albeit detectable, proinflammatory cytokine responses to S. pneumoniae, compared with alveolar macrophages from RelA(F/F) mice. Concentrations of these cytokines in lung homogenates were diminished early after infection, indicating a significant contribution of macrophage RelA to the initial expression of cytokines in the lungs. However, the cytokine content in infected lungs was equivalent by 15 hours. Neutrophil recruitment during S. pneumoniae pneumonia reflected a delayed onset in RelA(Δ/Δ) mice, followed by similar rates of accumulation. Bacterial clearance was eventually effective in both genotypes, but began later in RelA(Δ/Δ) mice. Thus, during pneumococcal pneumonia, only the earliest induction of the cytokines measured depended on transcription by RelA in myeloid cells, and this transcriptional activity contributed to effective immunity.

Figure 1.

Alveolar macrophages from RelA^Δ/Δ mice were present in comparable numbers, appeared similar in morphology, and expressed a nonfunctional form of RelA (missing exons 7–10) compared with alveolar macrophages from RelA^F/F littermates. (A) Morphology of macrophages recovered from mice did not differ between RelA^F/F (left) and RelA^Δ/Δ (right) mice, as indicated by Diff-Quik–stained cytocentrifuge preparations. (B) PCR amplification of normal and rearranged RelA (top) in DNA from alveolar macrophages of RelA^F/F and RelA^Δ/Δ mice demonstrated the gene to be rearranged in cells from the latter (Cre-positive) mice. PCR amplification of signal transducer and activator of transcription 3 (STAT3) (below) was used as a negative control to demonstrate the specificity of targeting. (C) Mutant but not wild-type RelA protein was detected in alveolar macrophages from Cre-positive RelA^Δ/Δ mice. Antibodies recognizing an epitope located within exons 7–10 (top) identified a product in the alveolar macrophages from RelA^F/F but not RelA^Δ/Δ mice. Antibodies recognizing a C-terminal epitope (middle) identified full-length and truncated forms of the RelA protein in RelA^F/F and RelA^Δ/Δ mice mice, respectively. β-actin antibodies (below) were used as a loading control.

Figure 2.

RelA was necessary for early-response cytokine production by alveolar macrophages stimulated with Streptococcus pneumoniae in vitro. Alveolar macrophages were collected from uninfected RelA^F/F or RelA^Δ/Δ mice and stimulated with S. pneumoniae for 2 hours. Medium was replaced every 2 hours after stimulation with S. pneumoniae, and concentrations of (A) TNF-α, (B) KC, and (C) macrophage inflammatory protein-2 (MIP-2) were measured in the supernatant. Data were expressed as mean ± SEM from three independent experiments. Data were analyzed using two-way ANOVA, followed by the Student Newman-Keuls post hoc test. *P < 0.05, compared with RelA^F/F mice.

Figure 3.

Production of inflammatory cytokines in alveolar macrophages depended on RelA during infection with S. pneumoniae. Quantitative RT-PCR was used to measure fold induction of (A) TNF-α, (B) IL-1α, (C) IL-1β, (D) KC, and (E) MIP-2 mRNA in alveolar macrophages collected by bronchoalveolar lavage from RelA^F/F and RelA^Δ/Δ mice infected with 10⁶ CFUs of S. pneumoniae for 6 hours. Data were expressed as geometric mean ± geometric SEM (n = 2–4 uninfected mice/group, or 4–9 infected mice/group). Data were analyzed using two-way ANOVA, followed by Bonferroni post hoc test. *P < 0.05, compared with RelA^F/F mice.

Figure 4.

Pattern recognition receptor expression in alveolar macrophages is partly dependent on RelA during infection with S. pneumoniae. Quantitative RT-PCR was used to measure fold induction of (A) Toll-like receptor 2 (TLR2), (B) Toll-like receptor 4 (TLR4), (C) scavenger receptor A (SR-A), and (D) macrophage receptor with collagenous structure (MARCO) mRNA in alveolar macrophages collected by bronchoalveolar lavage from RelA^F/F and RelA^Δ/Δ mice infected with 10⁶ CFUs of S. pneumoniae for 0 or 6 hours. Data were expressed as geometric mean ± geometric SEM (n = 6–9 uninfected mice/group, or 4–6 infected mice/group). Data were analyzed using two-way ANOVA, followed by Bonferroni post hoc test. *P < 0.05, compared with 0 hours for the same genotype. ^†P < 0.05, compared with RelA^F/F mice at same time point.

Figure 5.

Initial inflammatory cytokine protein production in the lungs during S. pneumoniae pneumonia was dependent on macrophage RelA. Concentrations of (A) TNF-α, (B) IL-1α, (C) IL-1β, (D) KC, and (E) MIP-2 in homogenized whole lungs from RelA^F/F and RelA^Δ/Δ mice infected with 10⁶ CFUs of S. pneumoniae for 0, 6, 15, or 48 hours were measured using ELISA. Data were expressed as mean ± SEM (n = 5–7 mice/group) and analyzed using two-way ANOVA, followed by Bonferroni post hoc test. *P < 0.05, compared with RelA^F/F mice at same time point after infection.

Figure 6.

Targeted deletion of RelA in myeloid cells impaired recruitment of neutrophils and compromised clearance of S. pneumoniae. (A) Representative sections of hematoxylin-and-eosin–stained lungs from uninfected or S. pneumoniae–infected (9 hours; 10⁶ CFUs) RelA^F/F and RelA^Δ/Δ mice, used for lung morphometry. (B) RelA^Δ/Δ mice had decreased neutrophils in their alveolar airspaces, measured using morphometry. Data were expressed as mean ± SEM (n = 9–11 mice/group). (C) Numbers of circulating neutrophils were not diminished in blood of RelA^Δ/Δ mice. (D) Bacterial clearance was compromised in RelA^Δ/Δ mice. Viable bacteria per lung were quantified using CFU assays, and data were expressed as mean ± SEM (n = 5–11 mice/group). Statistical analyses were performed using two-way ANOVA, followed by Student Newman-Keuls post hoc test. *P < 0.05, compared with RelA^F/F mice at the same time point after infection.