The pathomechanisms underlying the frequently observed fatal outcome of Klebsiella pneumoniae pneumonia in elderly patients are understudied. In this study, we examined the early antibacterial immune response in young mice (age 2–3 mo) as compared with old mice (age 18–19 mo) postinfection with K. pneumoniae. Old mice exhibited significantly higher bacterial loads in lungs and bacteremia as early as 24 h postinfection compared with young mice, with neutrophilic pleuritis nearly exclusively developing in old but not young mice. Moreover, we observed heavily increased cytokine responses in lungs and pleural spaces along with increased mortality in old mice. Mechanistically, Nlrp3 inflammasome activation and caspase-1–dependent IL-1β secretion contributed to the observed hyperinflammation, which decreased upon caspase-1 inhibitor treatment of K. pneumoniae–infected old mice. Irradiated old mice transplanted with the bone marrow of young mice did not show hyperinflammation or early bacteremia in response to K. pneumoniae. Collectively, the accentuated lung pathology observed in K. pneumoniae–infected old mice appears to be due to regulatory defects of the bone marrow but not the lung, while involving dysregulated activation of the Nlrp3/caspase-1/IL-1β axis.
This article is featured in Top Reads, p.
Elderly individuals aged >65 y are characterized by a greater susceptibility to bacterial and/or viral lung infections (1). Accordingly, increased morbidity and mortality rates are observed in this population with advanced age (2, 3). Similarly, old mice aged 18–20 mo develop increased lung inflammation, decreased lung function, and overall reduced survival in response to endotoxemia (4–6).
Dysregulated innate and adaptive immune responses to inflammation/infection characterize age-related poor prognosis and recovery from acute lung injury (7). For example, increased airway neutrophilia and serum levels of proinflammatory cytokines including IL-1β, IL-6, and TNF-α (senescence-associated secretory pathway) as well as decreased T and B cell proliferation characterize such “immunosenescence” (8, 9). As a consequence, immunosenescence restrains the immune system from adequately responding to microbial challenge, while at the same time providing a chronic proinflammatory milieu termed “inflamm-aging” (10). This feature may also add to the severity of adult respiratory distress syndrome in elderly patients (11). However, the molecular mechanisms underlying inflamm-aging remain insufficiently defined.
S. pneumoniae is the most prevalent Gram-positive pathogen in community-acquired pneumonia. However, lung infections with Gram-negative bacteria such as Haemophilus influenzae and Klebsiella pneumoniae increase in intensive care medicine, which is further complicated by the occurrence of antibiotic-resistant strains (12–14). Therefore, the risk of developing Gram-negative bacterial pneumonia in the elderly due to K. pneumoniae infection has increased substantially (15). Similar to humans, K. pneumoniae may also cause severe pneumonia in rodents (16, 17). Moreover, naturally acquired K. pneumoniae–induced pleuritis has been observed in California sea lions and Eurasian beaver (18, 19).
In the current study, we found that old but not young mice responded with early fibrinopurulent pleuritis, bacteremia, and increased mortality to K. pneumoniae infection of the lung. Moreover, increased Nlrp3/caspase-1/IL-1β activation characterized an inflamm-aging phenotype of K. pneumoniae–challenged old mice. Caspase-1 inhibition attenuated the hyperinflammation and bacterial outgrowth in K. pneumoniae–infected old mice. Finally, hematopoietic reconstitution of irradiated old mice with the bone marrow (BM) of young mice abolished the observed inflamm-aging phenotype of old mice. Possible interpretations of these findings are discussed.
Materials and Methods
Young and old male C57BL/6J mice were purchased from Janvier Laboratories (Le Genest-Saint-Isle, France) and were used in experiments at the age of 2–3 mo (young mice) and 18–19 mo (old mice). All mice were housed in individually ventilated cages and maintained under specific pathogen–free conditions with free access to food and water and were handled according to institutional guidelines of the Central Animal Facility of Hannover School of Medicine. Nlrp3−/− mice were purchased from The Jackson Laboratory (stock no. 021302) (20). All animal experiments were approved by the Lower Saxony State Office for Consumer Protection and Food Safety and followed the European Council Directive 2010/63/EU and the German Animal Welfare Act.
FITC- or allophycocyanin-conjugated anti-CD3 Ab (clone 145-2C11), PerCP-Cy5.5–conjugated anti-CD4 Ab (clone RM4-5), allophycocyanin-conjugated anti-CD8 Ab (clone 53-6.7), PE-Cy7–conjugated anti-CD45 Ab (clone 30-F11), FITC-conjugated TCRαβ Ab (clone H57-597), and PE-conjugated anti-TCRγδ Ab (clone GL-3) were all purchased from BD Biosciences (San Jose, CA). Recombinant murine GM-CSF (R&D Systems, Wiesbaden, Germany), IL-4 (PeproTech, Hamburg, Germany), and nigericin (specific Nlrp3 inducer) were purchased from InvivoGen. For Western blot analysis, monoclonal rabbit anti-mouse Nlrp3 Ab (clone D4D8T) was purchased from Cell Signaling Technology (Frankfurt, Germany). Monoclonal mouse anti-murine caspase-1 p20 Ab (clone Casper-1) was obtained from AdipoGen Life Sciences (San Diego, CA), and monoclonal mouse anti–β-actin Ab (clone AC-15) was purchased from Sigma-Aldrich (St. Louis, MO). Caspase-1 inhibitor Ac-YVAD-cmk was purchased from InvivoGen (San Diego, CA). DMSO was purchased from Carl Roth (Karlsruhe, Germany), and gentamicin was purchased from Sigma-Aldrich. Specificities of anti-Nlrp3 and anti–caspase-1 p20 Abs were confirmed by liquid chromatography–mass spectrometry of respective Western blot bands at the proteomics core facility of Hannover Medical School (21).
Culture and quantification of K. pneumoniae
K. pneumoniae (American Type Culture Collection strain 43816) was grown to midlog phase and snap-frozen in aliquots at −80°C. Aliquots of K. pneumoniae were plated on nutrient agar plates in 10-fold serial dilutions (BD Biosciences, Heidelberg, Germany). Subsequently, plates were incubated at 37°C, 5% CO2 for 18 h followed by determination of CFU (22–24).
Anesthesia and infection of mice with K. pneumoniae
Young and old mice were anesthetized with ketamine (75 mg/kg body weight; Albrecht, Aulendorf, Germany) and xylazine (5 mg/kg; Bayer, Leverkusen, Germany) followed by orotracheal intubation with a 26G Abbocath catheter (Abbott, Wiesbaden, Germany), as previously described (25, 26). Mice were infected with K. pneumoniae (106 CFU/mouse in 50 µl of PBS) and were then monitored daily for disease symptoms.
For caspase-1 inhibition, young and old mice were either treated with caspase-1–specific inhibitor Ac-YVAD-cmk (250 µg per 100 µl of DMSO/PBS i.p. [1:9, v/v]), or received 100 µl of DMSO/PBS (1:9, v/v) at 30 min, 6 h, and 12 h after K. pneumoniae infection, followed by collection of the various readouts, as described. In additional experiments, we verified that caspase-1 inhibitor Ac-YVAD-cmk did not affect bacterial growth itself by adding Ac-YVAD-cmk (200 and 400 µg/ml) or DMSO only (vehicle control) to Klebsiella pneumoniae in vitro (106 CFU in 1 ml of NB medium) followed by determination of CFU after 2, 4, and 6 h of incubation at 37°C/5% CO2.
Bronchoalveolar and pleural lavage and quantification of cellular constituents
Bronchoalveolar lavage (BAL) was performed as previously described (23, 24, 27, 28). Briefly, after inserting a shortened cannula into the trachea, repetitive instillations of 300-µl aliquots of ice-cold PBS/2 mM EDTA (Versen; Biochrom, Berlin, Germany) followed by subsequent aspiration were performed until a BAL volume of 1.5 ml was collected. After collection of an additional BAL fluid volume of 4.5 ml, both aliquots were centrifuged at 1400 rpm and 4°C for 9 min. Leukocytes were differentiated on Pappenheim-stained cytocentrifuge preparations, using overall morphological criteria, including cell size and shape of nuclei followed by quantification of the respective subsets by multiplication of total cell numbers with percent values of the respective leukocyte subsets. Resident alveolar macrophages were defined as cells with round or oval nuclei and abundant, even-shaped light-blue cytoplasm, whereas alveolar exudate macrophages recruited in response to infection were defined as cells showing one large and often kidney- or stellar-shaped nucleus with fewer and grayish-blue stained cytoplasm following Pappenheim staining. In addition, alveolar macrophages differ from exudate macrophages in view of their CD11chigh/CD11blow cell surface expression, whereas exudate macrophages are CD11chigh/CD11bhigh in flow cytometry analysis (24, 27, 29).
Pleural lavage (PL) was performed by inserting a 26G Abbocath catheter (Abbott, Wiesbaden, Germany) through the diaphragm into the pleural cavity under the right costal arch, and subsequent instillation of 500-µl aliquots of PBS into the pleural cavity followed by careful aspiration. The PL fluid (total 1.5 ml) was centrifuged at 1400 rpm at 4°C for 9 min followed by Pappenheim staining and subsequent differentiation of cellular constituents.
Determination of bacterial loads in BAL fluid, PL fluid, lung tissue, and peripheral blood
Bacterial loads were quantified in BAL fluid, PL fluid, lung tissue, and peripheral blood as previously described (21, 23, 27, 28). Briefly, BAL and PL fluid were sampled from mock- and K. pneumoniae–infected young and old mice, or old chimeric mice. For determination of CFU in lung tissue, individual lung lobes were removed, separated into pieces, and homogenized in 2 ml of HBSS (Merck, Darmstadt, Germany) using a tissue homogenizer (IKA, Staufen, Germany), followed by filtration through a 100-µm cell strainer (BD Biosciences). Blood was drawn from the vena cava inferior and collected into heparin-coated tubes (Sarstedt, Nümbrecht, Germany). Ten-fold serial dilutions of BAL fluid, PL fluid, lung tissue homogenates, and peripheral blood were plated on nutrient agar plates and incubated at 37°C, 5% CO2 for 18 h for subsequent enumeration of CFU.
Immunophenotypic analysis of leukocyte subsets in BAL fluid, PL fluid, and lung tissue
Flow cytometric analyses of leukocyte subsets in BAL fluid, PL fluid, and lungs were performed essentially as described previously (30, 31). Briefly, after collection of BAL fluids and PL fluids, lungs were minced into small pieces and digested in RPMI 1640 supplemented with collagenase A (5 mg/ml) and DNase I (1 mg/ml). CD45+ leukocytes were enriched from digested lung tissue using magnetic cell separation (MACS purification kit; Miltenyi Biotec, Bergisch Gladbach, Germany) (32). Flow cytometric analyses of leukocyte subsets were performed on a BD LSRFortessa (BD Biosciences) flow cytometer. Lymphocytes were gated according to their forward scatter area/side scatter area and CD45 cell surface expression. T cell subsets were gated by their CD45+/CD3+ and either CD4+ or CD8+ expression and were further examined for their differential αβ versus γδ TCR expression profiles. BD FACSDiva software was employed for postacquisition compensation of spectral overlaps of fluorescence channels and subsequent data analysis.
Cell sorting of lung neutrophils
FACS of lung neutrophils was performed with a BD FACSAria III cell sorter (Becton Dickinson, Heidelberg, Germany). Lungs of K. pneumoniae–infected young and old mice were minced into small pieces and digested in RPMI 1640 supplemented with collagenase A (5 mg/ml) and DNAse I (1 mg/ml). Ly6G+ cells were enriched from digested lung tissue using magnetic cell separation (MACS purification kit, Ly6G MicroBeads UltraPure, Miltenyi Biotec). Subsequently, neutrophils were gated according to their forward scatter area/side scatter area characteristics, followed by subgating on propidium iodide−, Ly6G+ cells. Sorting was performed at 4°C using a 70-µm nozzle tip. Postsort analysis of sorted neutrophils confirmed sort purities of >99%.
Western blot analysis of Nlrp3 and caspase-1 in PL cells and lung homogenates of K. pneumoniae–infected young and old mice
Nlrp3, pro–caspase-1, and active caspase-1 p20 protein expression was analyzed by Western blotting in PL cells and in lung tissue of young and old mice infected with K. pneumoniae for 16 h (21). Cells were lysed in ice-cold cell lysis buffer containing a mixture of protease inhibitors (aprotinin, leupeptin, PMSF, sodium orthovanadate; Sigma-Aldrich). In selected experiments, we verified that BAL and lung neutrophils did not demonstrate any residual neutrophil elastase (NE) activity after cell lysis (data not shown).
BM phagocytes from Nlrp3−/− mice served as a negative control for Nlrp3 analyses by Western blotting. To this end, Nlrp3-deficient phagocytes were cultured in DMEM (Life Technologies, Carlsbad, CA) supplemented with 10% FBS (Merck) and 100 U/ml penicillin/100 mg/ml streptomycin (Life Technologies) at 37°C and 10% CO2. For analysis of Nlrp3, pro–caspase-1, and active caspase-1 (p20), 5 × 105 BM phagocytes were seeded per well of a 48-well plate and were then primed with 1 µg/ml LPS for 4 h. Subsequently, cells were stimulated with 20 µM nigericin (serving as Nlrp3 activator) for 30 min followed by immunoblotting and detection of specific complex proteins according to recently published protocols (21). Quantification of Nlrp3 and active caspase-1 relative to β-actin was performed by using gel analysis software (BIO-1D; Vilber Lourmat, Eberhardzell, Germany). Equal numbers of lung neutrophils of K. pneumoniae–infected young and old mice (400,000 each) were lysed in ice-cold Western blot lysis buffer containing protease inhibitors followed by immunoblotting and subsequent detection of Nlrp3, pro–caspase-1, and active caspase-1 p20 protein expression according to recently published protocols (21).
Mock- or K. pneumoniae–infected young and old mice were euthanized at the indicated time points. Nonlavaged lungs were inflated with PBS-buffered formaldehyde solution (4.5%, pH 7, Roth) in situ and then removed en bloc for subsequent fixation for 24 h at room temperature, after which lungs were embedded in paraffin. Lung sections (3 µm) were prepared and stained with H&E. Evaluation of lung histopathology was performed under blinded conditions using an Olympus BX-53 microscope (Olympus, Tokyo, Japan).
Quantification of cytokines
Pro- and anti-inflammatory cytokines were measured in BAL fluids, PL fluids, and lung tissue supernatants by ELISA (R&D Systems) (detection limits for IL-1β, 12 pg/ml; TNF-α, 11 pg/ml; IL-6, 8 pg/ml; keratinocyte-derived cytokine (KC), 16 pg/ml; CCL2, 8 pg/ml; MIP-2, 8 pg/ml; and G-CSF, 14 pg/ml).
NE activity assay
NE activity in K. pneumoniae–infected young and old mice was examined by NE enzyme activity assay, essentially as described recently (33).
Generation of chimeric mice
To discern the effect of an aged hematopoietic system as compared with an aged lung on outcome in K. pneumoniae–infected old mice, in selected experiments, we reconstituted old mice with the hematopoietic system of young mice (chimeric young → old mice). To this end, aged donor mice were subjected to whole-body irradiation (8 Gy) followed by i.v. injection of BM cells (107 cells/mouse, tail vein injection) either collected from young mice or old mice (serving as BM transplantation [BMT] controls). For determination of successful BM engraftment, CD45.2 alloantigen-expressing old mice received BMT from CD45.1 alloantigen-expressing young mice. After irradiation and BMT, all mice were housed for at least 7 wk under individually ventilated cage conditions with free access to food and water. Hematopoietic engraftment of donor BM in recipient mice was always >92% (29, 34).
All data are shown as mean ± SD. Differences between groups were analyzed by a Mann–Whitney U test or unpaired t test, and survival data were compared using a log-rank test. Significant differences were assumed when p values were <0.05. Statistical analysis was performed with GraphPad Prism software (version 6).
Old but not young mice are susceptible to K. pneumoniae lung infection
K. pneumoniae is a Gram-negative pathogen known to cause suppurative pneumonia in mice and humans (35, 36). In this study, we examined the effect of age on outcome in K. pneumoniae pneumonia in mice. As shown in (Fig. 1, old mice exhibited significantly increased bacterial loads in all compartments examined, including BAL fluid as well as PL fluid, lung tissue, and peripheral blood as early as 24 h postinfection, when compared with young mice. Histopathological examination of lung tissue sections of young and old mice showed normal lung architecture in both groups at 24 h after mock infection (Fig. 2A, 2B). However, old but not young mice challenged with K. pneumoniae revealed dense neutrophilic infiltration of the lung interstitium and alveoli along with alveolar edema and intravascular coagulation (Fig. 2C, 2D). Of note, however, old mice very consistently developed fibrinopurulent pleuritis characterized by distinctive neutrophilic exfiltration as early as 24 h postinfection, which was not observed in PL fluids or lung histologies of K. pneumoniae–infected young mice (Fig. 2G–J). Moreover, a significantly increased NE activity accompanied such pleural neutrophilic exfiltrates in PL fluids of old but not young mice (Supplemental Fig. 1). In addition, old mice developed the histopathological pattern of pneumonia by 48 h postinfection, characterized mainly by neutrophil-dominated interstitial and alveolar leukocyte infiltration of the lung together with alveolar fibrinous exudate formation and intravascular coagulation (Fig. 2E, 2F). Overall, this histopathological pattern was accompanied by significantly increased mortality starting on day 1 postinfection in old as compared with young mice (Fig. 2K).
Next, we quantified inflammatory leukocyte recruitment in K. pneumoniae–infected old and young mice by flow cytometry. As shown in (Fig. 3, we found an increased cell recruitment in BAL and PL fluids of old as compared with young mice at 24, 48, and 72 h postinfection (Fig. 3A, 3B). Moreover, numbers of exudate macrophages and neutrophils but also lymphocytes were also significantly upregulated in BAL fluids of old mice at 24 and 48 h postinfection (Fig. 3C–F). In PL fluids, particularly neutrophil counts and, to a lesser extent, numbers of macrophages were increased in old as compared with young mice at 24 h up until 72 h postinfection (Fig. 3G, 3H). In addition, we found significantly increased numbers of CD4+ and CD8+ T cells in BAL fluid, lung tissue, and pleural spaces of aged as compared with young mice both under baseline conditions (CD8+ T cells in lung and pleura), and even more so in response to K. pneumoniae infection (Supplemental Fig. 2).
The Nlrp3/caspase-1/IL-1β axis contributes to K. pneumoniae pneumonia in aged mice
The Nlrp3/IL-1β axis is a critical inflammatory pathway in bacterial pneumonia (21). Therefore, we examined its contribution to the exacerbated pulmonary pathology observed in K. pneumoniae–challenged old mice. As shown in (Fig. 4A and 4B, we found a strongly increased Nlrp3 protein activity in both PL cells and lung homogenates of K. pneumoniae–challenged old mice compared with young mice. BM phagocytes from Nlrp3−/− mice stimulated with LPS and nigericin did not demonstrate a signal in Western blotting, thus indicating that the employed Western blotting Abs were specific for Nlrp3. In line with the upregulation of Nlrp3, we also found strongly increased pro–caspase-1 and active caspase-1 in PL cells but even more accentuated in lung tissue lysates of old as compared with young mice after challenge with K. pneumoniae, supported by quantification of Nlrp3, pro–caspase-1, and active caspase-1 protein expression relative to β-actin (Fig. 4C–F). We observed no increased Nlrp3 or caspase-1 activity in lungs of untreated young and old mice (Fig. 4B). Furthermore, neutrophils sorted from lungs of infected old mice exhibited a stronger Nlrp3 and active caspase-1 protein expression compared with the same number of lung neutrophils of infected young mice (400,000 cells each, Fig. 4G–I). These results indicate that the observed increased Nlrp3 and caspase-1 activation is not only due to increased lung neutrophil recruitment but is additionally affected by increased activation of recruited neutrophils in old mice upon Klebsiella pneumonia.
Because the Nlrp3/caspase-1 axis is required for IL-1β processing, we next measured IL-1β protein secretion in BAL fluids, lung tissue, and PL fluids of K. pneumoniae–challenged young and old mice. In old mice, we found significantly increased IL-1β protein secretion in PL fluids and lung tissue at 24 h and in BAL fluids at 48 h postinfection, relative to young mice (Fig. 4J, 4L, 4N). Other proinflammatory cytokines including TNF-α, IL-6, KC, and CCL2 were also increased in aged as compared with young mice at 24 and 48 h after K. pneumoniae infection (Fig. 4K, 4M, 4O, Supplemental Fig. 3A–I). We also observed significantly increased levels of G-CSF in plasma of old as compared with young mice after K. pneumoniae infection, and this G-CSF response was reduced in chimeric old mice (Supplemental Fig. 3J). Collectively, these data support the view that increased Nlrp3 inflammasome and caspase-1 activation together with enhanced IL-1β protein secretion contribute to the observed hyperinflammatory phenotype in old mice challenged with K. pneumoniae.
Effect of caspase-1 inhibition on K. pneumoniae challenge in old mice
We then aimed to confirm a central role for caspase-1 in mediating the hyperinflammatory response in old mice after challenge with K. pneumoniae. Therapeutic treatment of K. pneumoniae–infected old mice with caspase-1 inhibitor Ac-YVAD-cmk decreased IL-1β secretion in BAL fluids and lung tissue of K. pneumoniae–infected old mice (Fig. 5A, 5B). At the same time, we observed that several inflammatory variables, including bacterial loads (Fig. 5C–F) and infection-driven alveolar macrophage depletion (data not shown), were attenuated in inhibitor- relative to vehicle-treated K. pneumoniae–infected old mice. The same therapeutic treatment of K. pneumoniae–infected young mice with Ac-YVAD-cmk did not affect IL-1β secretion and bacterial killing (Fig. 5G–L).
Culture of K. pneumoniae in the presence of Ac-YVAD-cmk or DMSO only did not affect bacterial growth during an observation period of 6 h (Supplemental Fig. 4).
Old mice reconstituted with the hematopoietic system of young mice are resistant to K. pneumoniae pneumonia
Finally, we examined the effect of reconstitution of old mice with the hematopoietic system of young mice on lung-protective immunity against K. pneumoniae. As shown in (Fig. 6, chimeric old mice reconstituted with the BM from young donors (BMT young → old mice) demonstrated significantly less bacteremia and reduced bacterial loads in PL fluid and peripheral blood when compared with chimeric old mice reconstituted with BM from old mice (BMT old → old mice) at 24 h after K. pneumoniae infection (Fig. 6A–C). At the same time, this experimental maneuver led to reduced numbers of infiltrating neutrophils in PL fluids of chimeric old mice (BMT young → old mice), as determined by examination of Pappenheim-stained PL fluid cytospins (Fig. 6D, 6J, 6K). Moreover, we found significantly decreased proinflammatory IL-1β, TNF-α, and IL-6 protein secretion in PL fluids of K. pneumoniae–infected chimeric old mice (BMT young → old mice) (Fig. 6E–G). Neutrophil chemoattracting cytokines including MIP-2 and KC were also downregulated in this group (Fig. 6H, 6I). Histopathological examination of K. pneumoniae–challenged lungs from chimeric old mice confirmed a substantially reduced pleuritis, relative to transplantation controls (BMT old → old mice) (Fig. 6L, 6M).
In this study, we examined the early antibacterial response of the lung upon K. pneumoniae infection in young mice (aged 2–3 mo) as compared with old mice (aged 18–19 mo). We have identified several age-related effects contributing to the increased vulnerability of aged mice to Gram-negative bacterial infections. First, severe pleuritis was a key complication developing in old mice very early after K. pneumoniae challenge. Second, increased activation of the Nlrp3/caspase-1/IL-1β axis was noted in K. pneumoniae–infected old mice and was substantially attenuated by caspase-1 inhibitor treatment of old mice. Third, reconstitution with the hematopoietic system of young mice improved the lung pathology of old mice, suggesting that regulatory defects of the hematopoietic system contribute to the observed vulnerability of old mice upon K. pneumoniae infection. We believe that this study is of clinical interest because nosocomial pneumonia due to Gram-negative bacterial infections is increasingly observed in the elderly, with substantial lethality noted in this patient group. This situation is further complicated by the limited antibiotic availability within the pleural space. Future studies will need to address the question whether antibiotic-independent additive therapies (e.g., growth factor guided) will help to resolve senescence-associated impaired lung protective immunity in the elderly.
The data of the current study show an increased vulnerability of old mice to lung infection with K. pneumoniae. This study is consistent with previous reports also showing increased inflammation and reduced microbial killing in old as compared with young mice after Pseudomonas aeruginosa or Streptococcus pneumoniae challenge (37–39). However, we particularly observed a prominently developing pleuritis in K. pneumoniae–challenged old but not young mice. Naturally acquired K. pneumoniae infections have also been associated with severe pleuritis in mammalian species other than mice such as Eurasian beaver, Californian sea lions, and southern sea otters (18, 19). Corresponding literature on K. pneumoniae–induced pleuritis in elderly patients is sparse, most likely because this feature is difficult to detect in clinical intensive care (14).
Some studies showed that decreased mucous production and composition is an initial factor for age-related impaired lung protective immunity and disease progression in elderly mice (40, 41). Thus, we cannot exclude that neutrophilic pleuritis in K. pneumoniae–infected old mice may be favored, at least in part, by impaired mucociliary clearance. As such, age-related increased susceptibility to bacterial infection is a multifactorial process, and further studies are needed to clarify this aspect.
Particularly IL-1β was significantly increased in PL fluids and lungs of old mice upon K. pneumoniae challenge. IL-1β is a potent pleiotropic cytokine of the immediate early immune response after bacterial challenge and is mostly produced by monocytes/macrophages and neutrophils. The Nlrp3 inflammasome complex regulates the activation and processing of IL-1β (42). Its activation triggers caspase-1 to cleave pro–IL-1β into its active form for subsequent release of the cytokine (43). However, despite its important role in both acute and chronic inflammatory diseases (44–46), we and others recently found that excessive Nlrp3 activation and IL-1β secretion may aggravate the lung inflammatory response to bacterial challenge with fatal consequences for the infected host (21). Therefore, a tightly regulated inflammasome activation is central to successful pathogen eradication (21, 47). Telomere dysfunction as a consequence of aging has been associated with hyperactivation of the Nlrp3 inflammasome (48, 49). Other reports showed that transcription of inflammasome-component genes increases in blood mononuclear cells with age (50), and age-dependent Nlrp3 inflammasome hyperactivation was associated with increased lethality in elderly patients with SARS-CoV-2 pneumonia (51). Thus, dysregulated activation of the Nlrp3/caspase-1/IL-1β axis is a major driver of inflamm-aging and also appears to underlie the accentuated cytokine responses observed in K. pneumoniae–infected old mice. This concept is supported by our observation showing that caspase-1 inhibition in K. pneumoniae–infected old mice significantly decreased lung IL-1β protein levels and bacterial loads in K. pneumoniae–infected old mice. Therefore, caspase-1 inhibition may be an important adjunctive therapy for elderly patients suffering from severe K. pneumoniae lung infection.
Another study demonstrated age-associated immune impairments regarding the RIG-1 signaling pathway and Nlrp3 inflammasome activation in elderly female BALB/c mice with nasal-induced influenza virus infection, whereas therapeutic treatment with ATP and nigericin led to a reactivation of these pathways (52). In the current study, we found an increased rather than decreased activation of the Nlrp3/caspase-1 axis during Gram-negative bacterial pneumonia in old male C57BL/6J mice, which were mitigated by specific caspase-1 inhibition. Therefore, while considering differences in pattern recognition receptor pathways between bacterial and viral pathogens, Nlrp3 inflammasome activation patterns may also differ in elderly mice depending on the pathogen under study.
Transplantation of young but not old BM cells into irradiated old mice substantially ameliorated the observed severe lung pathology of old mice. These experiments suggest that not the old lung itself is incapable of self-limiting inflammation upon bacterial challenge, but rather age-related inflammatory dysregulation originating from the hematopoietic system appeared to contribute to the observed lung inflammatory response in old mice. As a consequence, aging of the hematopoietic system has a negative “remote” effect on the lung inflammatory response to K. pneumoniae infection. This interpretation is supported by other studies showing beneficial effects of young BMT to restore age-related impaired cardiac angiogenesis in old mice (53), as well as to alleviate renal aging in old mice (54). Future studies will need to specify age-related differences between young and old BM cellular constituents to develop therapeutic approaches to overcome immunosenescence and/or inflamm-aging-related clinical disorders.
In summary, we show that K. pneumoniae triggers pleuritis, severe bacteremia, and early mortality in old but not young mice. This phenotype is characterized by increased activation of the Nlrp3/caspase-1/IL-1β axis, and is attenuated by caspase-1 inhibitor treatment of old mice challenged with K. pneumoniae. Transplantation of young BM cells into old mice significantly ameliorated the pathology of K. pneumoniae pneumonia in old mice. This observation strongly suggests that senescence of the hematopoietic system and not the lung itself underlies the observed lung pathology in K. pneumoniae–infected old mice. As such, our study provides novel insights into age-related differences between young and old mice and their lung responses to nosocomial pathogens. The findings may serve as a basis for the development of new therapeutic interventions to ameliorate age-related disorders in elderly patients.
We thank Prof. Andreas Pich from the proteomics core facility of Hannover Medical School for confirming specificities of anti-Nlrp3 and anti–caspase-1 p20 Abs by liquid chromatography–mass spectrometry analysis.
This work was supported by the Deutsche Forschungsgemeinschaft (to C.B. and U.A.M.) and by the Bundesministerium für Bildung und Forschung for funding the German Center for Lung Research. Financial support was also provided by the Lower Saxony Society for the treatment of pulmonary diseases.
The online version of this article contains supplemental material.
The authors have no financial conflicts of interest.