|Home | About | Journals | Submit | Contact Us | Français|
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 (RelAΔ/Δ) with littermate controls (RelAF/F). Alveolar macrophages from RelAΔ/Δ mice expressed no full-length RelA, demonstrating effective targeting. Alveolar macrophages from RelAΔ/Δ mice exhibited reduced, albeit detectable, proinflammatory cytokine responses to S. pneumoniae, compared with alveolar macrophages from RelAF/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 RelAΔ/Δ mice, followed by similar rates of accumulation. Bacterial clearance was eventually effective in both genotypes, but began later in RelAΔ/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.
Acute lung infections cause more disease worldwide than neoplasms, ischemic heart disease, or HIV/AIDS. However, their importance to public health is relatively underrepresented as a research priority (1). Bacterial pneumonias, a common cause of morbidity and mortality in the United States and worldwide, are estimated to cost the United States approximately $20 billion per year (2). The most common cause of community-acquired pneumonia is the Gram-positive bacterium Streptococcus pneumoniae (3), commonly called pneumococcus.
The recruitment of neutrophils into infected airspaces is crucial for the clearance of invading pneumococci (4, 5). This process is largely made possible by chemokines in which 2 amino-terminal cysteines are separated by a single amino acid (CXC) and preceded by a glutamic acid-leucine-arginine (ELR) motif including KC (CXCL1) and microphage inflammatory (MIP)-2 (CXCL2) in mice, which can be expressed by alveolar macrophages and alveolar epithelial cells (6, 7). During pneumococcal pneumonia, the recruitment of neutrophils and the expression of ELR+ CXC chemokines require upstream signaling from the early response cytokines, TNF-α and IL-1 (specifically, IL-1α and IL-1β) (8).
The induction of TNF-α, IL-1, ELR+ CXC chemokines, and other inflammatory cytokines is mediated by multiple transcription factors, including NF-κB. Of the five NF-κB proteins, only p50 and RelA (also known as p65) are readily detectable in lung nuclear fractions during acute pulmonary inflammation (8–10). p50 limits the expression of inflammatory cytokines and prevents inflammatory lung injury (11, 12). In contrast, RelA drives proinflammatory responses by promoting the expression of inflammatory cytokines, and the deletion of RelA from all cells severely compromises antibacterial host defense (13, 14). Distinct roles of RelA in different lung cell types during pneumococcal pneumonia remain to be discriminated.
During bacterial pneumonias, alveolar macrophages are the first leukocytes to encounter pathogens (15, 16). Based on studies using cytotoxic agents to diminish the number of alveolar macrophages, these cells were observed to play roles in eliminating bacteria (17–19), facilitating neutrophil migration (20), and resolving inflammation (21, 22). Macrophage-derived cytokines were inferred to be critical in lung defense, and these cytokines are dependent on NF-κB (23), but this inference has not been rigorously tested. The deletion of RelA in myeloid cells leads to decreased levels of cytokines in the lungs after the instillation of a bolus of heat-killed Pseudomonas aeruginosa (24), suggesting that the activity of RelA in macrophages may be involved in net cytokine elaboration during lung infection. In contrast, the deletion of the NF-κB–activating enzyme IkB Kinase b (IKKb) from myeloid cells leads to the excess elaboration of cytokines from macrophages in response to LPS or Gram-positive bacteria (25, 26), suggesting that IKKβ and perhaps NF-κB are important in macrophages for limiting rather than promoting the expression of cytokines and inflammatory responses. Because alveolar macrophages are sentinel cells recognizing pneumococcus in the lungs (27), and because RelA is necessary for the expression of cytokines mediating the recruitment of neutrophils (13, 14), we hypothesized that alveolar-macrophage NF-κB RelA is critical for the initiation of the innate immune response during pneumococcal pneumonia.
Mice lacking RelA in myeloid cells were generated by crossing LysM-Cre mice (28) with Rela-floxed mice (29), generating a colony producing LysM-Cretg+/RelaloxP/loxP (RelAΔ/Δ) and LysM-Cretg−/RelaloxP/loxP (RelAF/F) littermates. Fifty microliters of saline containing S. pneumoniae serotype 19 (SP19; strain EF3030, kindly provided by Dr. Marc Lipsitch, Boston, MA) or serotype 3 (SP3; strain 6303; ATCC, Manassas, VA) were instilled into the left bronchus of 6–12-week-old mice, as described elsewhere (13). Our protocols were approved by the Institutional Animal Care and Use Committees at Harvard University and Boston University.
Lungs were lavaged 10 times, as described previously (30), with 1 ml of ice-cold buffer (20 mM Hepes, 145 mM NaCl, 5.5 mM dextrose, 2.7 mM EDTA, and 100 U/ml Pen-Strep; Invitrogen Life Technologies, Carlsbad, CA). Cells were washed twice with saline. Protein lysates were collected as described elsewhere (13). DNA or RNA was isolated using the DNeasy blood and tissue kit (Qiagen, Valencia, CA) or TRIzol reagent (Invitrogen Life Technologies, Valencia, CA), respectively.
Recombination of the Rela locus in alveolar macrophages from RelAΔ/Δ mice was confirmed, using PCR with the following three primers: 5′-CCAGCGAATCCAGACCAACAATAAC-3′, 5′-CGACCACTCTTCACCTGATTGTTTTTC-3′, and 5′-AAGGCTCAGAGGCAGTGAGAGACATAC-3′. Amplification of full-length or rearranged Rela was indicated by PCR products of approximately 2,500 or 1,600 bp in length, respectively.
RelA protein was assessed in cell lysates by immunoblotting (10, 13), using polyclonal antibodies raised against RelA (Cell Signaling Technology, Danvers, MA, or Santa Cruz Biotechnology, Santa Cruz, CA) or β-actin (Cell Signaling Technology).
Lavaged cells were cultured overnight in serum-free RPMI-1640 (Invitrogen Life Technologies) with penicillin and streptomycin. After 1 hour, nonadherent cells were removed by washing. Cells were washed twice, and bacteria were added in media without antibiotics. After 2 hours, the supernatants were collected, and the cells were washed twice and cultured in media containing antibiotics. Media were replaced and supernatants were collected every 2 hours.
mRNA induction for TNF-α, IL-1α, IL-1β, KC, MIP-2, Toll-like receptor 2 (TLR2), TLR4, scavenger receptor A (SR-A), and macrophage receptor with collagenous structure (MARCO) were measured in left lobes or bronchoalveolar lavage (BAL) cells, using quantitative RT-PCR (13). Primer and probe sequences are listed in Table 1 or published elsewhere (13, 31, 32). Cytokine protein concentrations were measured in culture supernatants or whole-lung tissue homogenates (30), using Quantikine ELISA kits (R&D Systems, Minneapolis, MN).
Alveolar neutrophils were quantified by morphometric analyses of hematoxylin-and-eosin–stained sections of left (infected) lung lobes, as described previously (30, 33, 34). Peripheral blood from the vena cava was used for total and differential cell counts.
Lungs were homogenized in sterile, distilled H2O containing protease inhibitors (35). Homogenates were serially diluted, plated on 5% sheep blood agar plates using the drop plate method (36), and incubated overnight at 37°C. Colonies were then counted to quantify CFUs/lung.
Statistics were calculated using GraphPad Prism (GraphPad Software, La Jolla, CA) and/or Statistica (StatSoft, Tulsa, OK). Data were analyzed using unpaired Student t tests for comparisons between two individual groups or two-way ANOVA, followed by post hoc tests, for individual differences between multiple groups, as indicated. P < 0.05 was considered significant.
Although Rela−/− mice die during gestation (37), RelAΔ/Δ and RelAF/F mice were born in numbers reflecting Hardy-Weinberg equilibrium, indicating that the deletion of RelA in myeloid cells alone does not confer a selective disadvantage during gestation. Rela−/− mice survive through birth if they lack tumor necrosis factor receptor 1 (TNFR1) signaling, but display severely shortened life spans because of bacterial infections (14). In contrast, mice with RelA mutated in myeloid cells alone survived to weaning, and demonstrated no gross signs of morbidity through at least 30 weeks of age.
To determine whether RelA was effectively targeted in alveolar macrophages, we collected alveolar macrophages from RelAΔ/Δ and RelAF/F mice, and isolated their DNA and proteins for RelA analyses. No significant differences were evident in the numbers (data not shown) or morphology (Figure 1A) of alveolar macrophages collected from RelAΔ/Δ and RelAF/F mice. PCR with primers specific for the wild-type Rela gene or the rearranged gene product (missing exons 7–10) demonstrated that Rela was rearranged in DNA from the alveolar macrophages of RelAΔ/Δ mice, but not from those of RelAF/F mice (Figure 1B). We used two different RelA-specific antibodies to determine whether truncation of the RelA protein occurred in alveolar macrophages from RelAΔ/Δ mice. One antibody (catalogue number 3034; Cell Signaling Technology) recognizes the amino acids surrounding Ser276 of human NF-κB RelA, encoded by exons excised by Cre recombinase if targeting is effective as designed. Alveolar macrophages from RelAΔ/Δ mice had no detectable RelA according to this antibody (Figure 1C, top). The other RelA antibody (catalogue number sc-372; Santa Cruz Biotechnology) recognizes the carboxy terminus of RelA, encoded by exons outside the loxP targeting strategy. This epitope should be present in both wild-type (full-length) and mutant (truncated) forms of RelA. The antibody demonstrated that alveolar macrophages from RelAΔ/Δ mice expressed only the truncated version of RelA, whereas those from RelAF/F mice expressed only the full-length RelA (Figure 1C, middle). Thus, alveolar macrophages are effectively targeted by the LysM-Cre transgene. Alveolar macrophages in RelAΔ/Δ mice express no full-length RelA.
To determine whether alveolar macrophages produce cytokines in response to S. pneumoniae and whether RelA is necessary for this production, we isolated alveolar macrophages from RelAΔ/Δ and RelAF/F mice and cultured them in vitro with SP19 for 2 hours. Supernatants were removed, and the medium was replaced every 2 hours. Alveolar macrophages from control RelAF/F mice produced TNF-α, KC, and MIP-2 in response to SP19 infection in vitro (Figure 2). This cytokine expression was transient, with the majority of expression occurring within 4 hours after initial exposure of macrophages to bacteria. Between 4–6 hours, little to no TNF-α, KC, or MIP-2 was expressed (Figure 2), and none was detectable after 6 hours of culture (data not shown). Neither IL-1α nor IL-1β was detected in these supernatants from macrophages infected in vitro with SP19 (data not shown), which may reflect their distinctive mechanisms of secretion (38, 39). Alveolar macrophages from RelAΔ/Δ mice expressed significantly less TNF-α and MIP-2 in response to infection with SP19 than alveolar macrophages from RelAF/F mice after infection with SP19 (Figure 2), and KC showed a similar trend that did not reach statistical significance. Alveolar macrophages from RelAΔ/Δ mice still expressed some TNF-α, KC, and MIP-2 (Figure 2), suggesting that additional pathways mediating the RelA-independent expression of these cytokines exist. However, the data suggest that the predominant pathway for the expression of these proinflammatory cytokines from isolated alveolar macrophages stimulated with pneumococcus requires functional RelA.
To determine whether the expression of alveolar macrophage cytokines is RelA-dependent during pneumococcal pneumonia, we instilled SP19 into the lungs of RelAΔ/Δ and RelAF/F mice and collected alveolar macrophages by BAL 6 hours later. At this early time point of infection, which likely included cells from both infected and uninfected lung regions, over 95% of the collected cells were macrophages. During these in vivo infections, alveolar macrophages from RelAΔ/Δ mice had significantly less mRNA for TNF-α, IL-1α, IL-1β, KC, and MIP-2 compared with alveolar macrophages from RelAF/F mice, indicating that the induction of these cytokines in alveolar macrophages is largely RelA-dependent during pneumococcal pneumonia. However, some TNF-α, IL-1α, IL-1β, KC, and MIP-2 mRNA was produced by these cells (Figure 3). Together with the in vitro analyses, these data from in vivo infections indicate that pneumococcus induces the expression of inflammatory cytokines and chemokines by alveolar macrophages via primarily RelA-dependent pathways.
To determine whether RelA influenced the expression of macrophage receptors for pneumococcus, we measured relevant transcripts in alveolar macrophages lavaged from control and mutant lungs with and without pneumonia. Pneumococcus can be recognized by TLR2 (40), TLR4 (41), SR-A (42), and MARCO (27). In control RelAF/F mice, pneumococcus increased the expression of TLR2, SR-A, and MARCO, but not TLR4 (Figure 4). In contrast, none of these receptors were induced by pneumococcal exposure in RelAΔ/Δ mice (Figure 4), demonstrating that increased responsiveness to pneumococcus is facilitated by this transcription factor in these cells. Significantly less MARCO mRNA was evident in RelAΔ/Δ macrophages compared with RelAF/F macrophages even before infection, and less TLR2 and MARCO mRNA during infection, indicating that the initial recognition of pneumococcus by alveolar macrophages may be compromised by the targeting of RelA. These data suggest that the recognition by alveolar macrophages of pneumococcus is partly dependent on a positive feedback loop involving pattern recognition receptors and RelA.
To determine the degree to which macrophage RelA contributes to the total lung content of these cytokines during pneumococcal pneumonia, we measured protein concentrations of TNF-α, IL-1α, IL-1β, KC, and MIP-2 in lung homogenates from RelAΔ/Δ and RelAF/F mice at 0, 6, 15, and 48 hours after infection with SP19. Before infection with S. pneumoniae, no significant differences were evident in lung cytokine concentrations between RelAΔ/Δ and RelAF/F mice (Figures 5A–5E). Infection increased the lung content of all cytokines measured (Figure 5), likely because of induction in the infected sites, although both infected and uninfected lung regions were included in these analyses. Early after infection (6 hours), the expression of cytokines was significantly reduced in whole lungs of RelAΔ/Δ mice because of a macrophage RelA deficiency. At later times after infection (up to 48 hours), total lung cytokine concentrations were no longer decreased in the lungs of mice lacking RelA in their macrophages (Figures 5A–5E). Altogether, these data indicate that macrophage RelA is particularly important for the early elaboration of these cytokines during pneumococcal pneumonia.
Because the early-response cytokines and ELR+ CXC chemokines that depend on macrophage RelA contribute to neutrophil recruitment in the lungs (23), we hypothesized that RelAΔ/Δ mice would have fewer emigrated neutrophils compared with RelAF/F littermates during pneumonia. By 9 hours after infection with SP19, neutrophils were evident in the airspaces of RelAF/F mice (Figures 6A and 6B). At this time, RelAΔ/Δ mice had significantly fewer neutrophils in their alveoli than did the RelAF/F mice (Figures 6A and B), at 32% of control levels. Although neutrophils in the alveolar airspaces increased over time in both genotypes, neutrophil recruitment in the RelAΔ/Δ mice remained lower than in RelAF/F mice throughout the entire 48-hour time course (Figure 6B). Reduced numbers of neutrophils in the alveoli counts were not secondary to neutropenia, because RelAΔ/Δ mice had as many or more circulating neutrophils as RelAF/F mice throughout their pneumonia (Figure 6C). These data suggest that, in addition to its influence on cytokine and receptor synthesis (Figures 2–5),), macrophage RelA contributes to maximal neutrophil recruitment into the alveolar airspaces during pneumococcal pneumonia.
Because neutrophil recruitment mediates bacterial clearance during pneumococcal pneumonia (4, 5) and was impaired in RelAΔ/Δ mice, we measured viable bacteria in the lung to determine whether myeloid RelA plays roles in integrated host defense. All lung lobes were included together in these analyses; hence the numbers of CFUs recovered represent all bacteria throughout the lower respiratory tract. At 6 hours, before the substantive recruitment of neutrophils into the airspaces, lung bacterial burdens were increasing, with no difference attributable to genotype (Figure 6D). Coinciding with neutrophil recruitment in RelAF/F mice, bacterial burdens plateaued, followed by a decline. By 48 hours after pneumococcal infection, RelAF/F mice had cleared bacteria to levels below the initial inoculum of 106 CFUs (Figure 6D). By 15 hours after pneumococcal infection, the lungs of RelAF/F mice contained significantly fewer living bacteria than those of RelAΔ/Δ mice, indicating an important role for myeloid RelA in this bacterial clearance (Figure 6D). RelAF/F mice continued to contain fewer S. pneumoniae in their lungs through 48 hours after infection compared with RelAΔ/Δ mice (Figure 6D). Altogether, these data suggest that alveolar macrophage RelA is necessary for optimal bacterial clearance during pneumococcal pneumonia.
As previously reported (43), the infections performed with serotype 19 pneumococcus were self-limiting. Both genotypes of mice cleared the organisms and survived. To determine whether the early defect in macrophage-derived cytokines influences the outcome of a more severe infection, mice were challenged with higher doses of this bacteria (3 × 107 instead of 1 × 106 CFUs per mouse), sufficient to cause bacteremia and death (43). During high-dose infection, no differences were evident between genotypes in terms of bacteremia (two of six in each genotype at 48 hours) or mortality (four of six in each genotype at 7 days). We also addressed a model of severe infection that provided a means of extending observations to a different bacterium, a serotype 3 pneumococcus, which is extremely virulent at even low levels of infection (8, 43). As expected, based on previous studies of this organism in mice with compromised innate immunity (8), the growth of this virulent bacterium in the lungs was unaltered by murine genotype (Table 2). However, when cytokines and neutrophils were measured in the lungs of mice with serotype 3 infection, the myeloid RelA mutation showed phenotypes similar to those in the less virulent serotype 19 infection (Table 2), including the effects of myeloid RelA targeting on both the expression of TNF-α and the accumulation of neutrophils. These data suggest that myeloid RelA contributes to early innate immune responses during diverse infections, but that RelA-dependent macrophage responses may be especially important for infections that are less severe because of either the dose or virulence of the microbe.
These studies identify roles of alveolar macrophage RelA in cytokine and receptor expression initiating innate immune responses during pneumococcal pneumonia. RelA is necessary in alveolar macrophages for the induction of cytokines and chemokines elicited by pneumococcus. During pneumonia, macrophage RelA contributes greatly to the early burst of cytokines, with long-term consequences for neutrophil recruitment and bacterial clearance. Thus, RelA-mediated cytokine expression by alveolar macrophages has important proinflammatory functions during pneumococcal pneumonia.
The alveolar macrophage expression of cytokines is largely but not completely dependent on RelA during pneumococcal infection. In in vitro studies with purified cultures of alveolar macrophages, the vast majority of cytokine elaboration stimulated by pneumococcus required RelA. When alveolar macrophages were collected from murine lungs infected in vivo with pneumococcus, their cytokine expression was nearly abrogated by the deletion of functional RelA. Consistent observations across in vitro experiments (in which cultured cells are not exposed to potentially relevant host factors) and in vivo experiments (in which macrophages are in a dynamic environment, beyond the control of investigators) inspire confidence that these results effectively depict the role of RelA in alveolar macrophages. Because we observed a decreased expression of TLR2 and MARCO in RelA mutant macrophages during pneumonia, signaling other than that downstream of RelA may conceivably be compromised because of the RelA mutation. Hence, the total effect of RelA mutation on cytokine expression may include both direct and indirect effects. However, at least some expression of cytokines is clearly independent of alveolar macrophage RelA. The low-level expression of cytokines remaining in alveolar macrophages without RelA may be mediated by transcription factors such as Ets, Elk-1, Egr-1, ATF-2, c-Jun, Sp1, PU.1, and IRF-4, each of which can function in the transcription of TNF-α and IL-1β (44, 45). The present study, however, demonstrates that in response to pneumococcus, RelA is particularly important to the expression of these cytokines by alveolar macrophages.
When whole-lung lysates were examined, all cytokines were significantly reduced in RelAΔ/Δ mice at the 6-hour time point. The model of LysM-targeted deletion does not allow for discrimination between RelA function in alveolar macrophages and other cells of the myeloid lineage such as neutrophils and dendritic cells, all of which are represented at some concentration in whole-lung lysates. However, alveolar macrophages collected from the airspaces of infected mice expressed cytokines, and these cytokines were diminished by RelA disruption. Thus, the earliest cytokine elaboration during pneumococcal pneumonia is dependent on RelA in myeloid cells (likely alveolar macrophages). We conclude that during pneumococcal pneumonia, the initial, early burst of cytokine expression in the lungs requires RelA in alveolar macrophages.
Interestingly, by 15 hours of infection, no significant differences attributable to genotype were evident in the lung homogenate concentrations of most cytokines. Because alveolar macrophage cytokines are so dependent on RelA, cells other than macrophages are likely sources of these cytokines during later stages of pneumococcal pneumonia. Alveolar epithelial cells, airway epithelial cells, airway smooth muscle cells, interstitial fibroblasts, microvascular endothelial cells, and lung lymphocytes can all generate cytokines that may contribute to inflammatory signaling in infected lungs (23). The delayed time for other cells to produce cytokines may result from changes in bacterial physiology during infection. For example, over time, pneumococci in the lungs shed their capsules and physically interact with structural cells in the lung (46), and the NF-κB responses of epithelial cells to pneumococcus positively correlate with bacterial adherence (47). Such changes in interaction between lung cells and pneumococcus may stimulate the expression of cytokines from cells other than macrophages during in vivo infections. Moreover, RelA-independent signaling may promote the expression of cytokines by macrophages in RelAΔ/Δ mice at later stages of pneumonia. Our results indicate that purified alveolar macrophages express some cytokines even in the absence of functional RelA, albeit to a much lower extent. The inflammatory milieu of a pneumonic lung by 15 hours may be sufficiently different to promote significant synthesis of cytokines downstream from alternative signaling pathways in cells lacking RelA. Lastly, the effect of increased lung bacterial burden evident after 15 hours of infection in RelAΔ/Δ mice may provide a more robust stimulus for the expression of cytokines in the lungs, superseding the observed negative effect of macrophage RelA deletion on this response.
The defect in cytokine expression from RelA-deficient macrophages significantly compromised the integrated host defense of the lung, diminishing the recruitment of neutrophils and exacerbating pneumococcal infection. The clearance of bacteria appeared to be delayed rather than prevented in mutant mice. Similarly, the recruitment of neutrophils appeared to be staggered. It was slower to start in the mutant mice, and then neutrophils accumulated at a comparable rate in both genotypes, resulting in lower numbers of neutrophils in the mutants than in control mice throughout the time course measured. The kinetics of both these host defense functions are consistent with a lag effect, in which a transient defect in innate immunity in mutant mice early during infection is followed by competent responses. This interpretation again emphasizes the concept that macrophage RelA is particularly important in the initial responses to microbes in the lungs.
Defects in host defense were also evident in mice with LysM-targeted RelA deletion during P. aeruginosa pneumonia (24), but the host defense phenotype during Group B Streptococcus (GBS, S. agalactiae) pneumonia in mice with a LysM-targeted deletion of IKKβ was dramatically different (25). The deletion of IKKβ in macrophages resulted in an increased recruitment of neutrophils and decreased bacterial burdens during GBS pneumonia, attributed to enhanced expression of IL-12 (25). We did not observe any increase in lung IL-12 because of RelA targeting in macrophages during pneumococcal pneumonia (data not shown). The deletion of IKKβ in macrophages also led to death during endotoxemia, attributable to increased cytokines driven primarily by an elevated expression of IL-1β (26). In contrast, our studies with LysM-targeted RelA deletion demonstrated decreased cytokines, including IL-1β, at early time points of pneumococcal pneumonia, and at no time did we observe increased IL-1β or increased inflammatory responses. LysM-targeted RelA deletion also decreased rather than increased BAL fluid IL-1β after the instillation of heat-killed P. aeruginosa (24). IKKβ is a kinase with many substrates that may influence many signaling pathways, apart from or in parallel with the activation of RelA (48). Although differences in study design may contribute to the disparate results already described, we suggest that IKKβ in macrophages may limit the inflammatory responses elicited by bacterial products through its influence on pathways other than the activation of NF-κB RelA.
Although our studies clearly indicate roles for RelA in mediating the expression of cytokines from alveolar macrophages, integrated host responses such as bacterial clearance and neutrophil recruitment may also result from additional effects of RelA mutations in cells other than alveolar macrophages. The LysM-targeting strategy used here targeted neutrophils and some dendritic cells (49). Because RelA mutations in alveolar macrophages clearly decrease the expression of TNF, IL-1, KC, and MIP-2 from alveolar macrophages, and interruptions of these signaling pathways diminish the recruitment of neutrophils and bacterial clearance in the lungs (8, 50), we favor the interpretation that the early loss of alveolar macrophage–derived cytokines is responsible for the early defects in neutrophil recruitment and bacterial clearance.
Lastly, these studies focused exclusively on innate immunity, and the cytokines implicated here as dependent on myeloid RelA early during infection have clear roles in orchestrating the innate immune defenses of the lung (23). In addition, myeloid-derived cytokines such as IL-1β can influence adaptive immune responses (51), and adaptive immunity can provide effective protection against pneumococcal infections (52). The influences of myeloid RelA on adaptive immune defenses against pneumococcus, such as antibody production or Th17 skewing, remain to be determined.
Our studies suggest that alveolar macrophage RelA plays a crucial role in the innate immune response during pneumococcal pneumonia by initiating the expression of important inflammatory cytokines and chemokines. Inhibiting this early cytokine expression by targeting RelA in macrophages exerts long-lasting effects on the recruitment of neutrophils and bacterial clearance during pneumococcal pneumonia. Based on these observations, we anticipate that natural variations or pharmacologic manipulations that decrease macrophage RelA activity may predispose individuals to pneumococcal pneumonia. The results also offer clear evidence for the expression of cytokines occurring independent of macrophage RelA during pneumococcal pneumonia. The cellular sources, molecular mechanisms, and physiologic significance of this macrophage RelA-independent cytokine elaboration constitute important avenues for future research.
The authors thank Dr. Marc Lipsitch for providing S. pneumoniae serotype 19, Dr. Peter Mancuso for guidance in alveolar macrophage culture, and Amber Thomas for technical assistance. The authors also thank Dr. Matthew Jones and Matthew Blahna for their scientific input.
This work was supported by National Institutes of Health grants HL068153 (J.P.M.), HL079392 (J.P.M.), HL092956 (L.J.Q.), and HL092743 (L.A.P.). L.J.Q. was supported by an American Lung Association Senior Research Fellowship.
Originally Published in Press as DOI: 10.1165/rcmb.2010-0210OC on January 14, 2011
Author Disclosure: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.