Adenosine receptor A3 is a critical mediator in LPS-induced pulmonary inflammation

Rosalyn Wagner, Kristian-Christos Ngamsri, Stefanie Stark, Irene Vollmer, Jörg Reutershan


Adenosine receptor A3 (A3) regulates directed movement of polymorphonuclear cells (PMNs) to sites of inflammation and has been implicated as a relevant mediator in models of inflammatory diseases. Here, we sought to characterize the role of A3 in a murine model of lung inflammation. Initial studies revealed that pulmonary A3 transcript levels were elevated following LPS exposure in vivo. In addition, inhalation of LPS increased the accumulation of PMNs in wild-type and A3−/− mice in all lung compartments. Pretreatment with the specific A3-agonist Cl-IB-MECA significantly decreased migration of PMNs into lung interstitium and alveolar air space of wild-type mice but not of A3−/− mice. Lower PMN counts were associated with reduced levels of TNF-α and IL-6 in the alveolar space of wild-type mice that received Cl-IB-MECA. In addition, Cl-IB-MECA attenuated LPS-induced microvascular permeability in wild-type mice as assessed by the extravasation of Evans blue. In pulmonary microvascular endothelial cells, Cl-IB-MECA reduced LPS-induced cytoskeletal remodeling and cell retraction, consistent with a specific role of A3 for maintaining endothelial integrity. Migratory activity of human PMNs across an endothelial or epithelial monolayer was reduced when A3 was activated on PMNs. Studies in chimeric mice, however, revealed that Cl-IB-MECA required A3 on both hematopoietic and nonhematopoietic cells to reduce transmigration in vivo. Together, our results shed new light on the role of A3 in LPS-induced PMN trafficking in the lung and suggest pharmacological modulation of A3-dependent pathways as a promising approach in lung inflammation.

  • acute lung injury
  • polymorphonuclear leukocytes
  • pulmonary circulation
  • adenosine receptors

acute lung injury (ALI) is a life-threatening disorder with significant adverse impact on morbidity and mortality of critically ill patients. As recent epidemiological studies indicate, ALI leads to 75,000 deaths annually in the United States (47). Respiratory failure is caused by an excessive inflammatory response to both pulmonary and extrapulmonary stimuli, including pneumonia, acid aspiration, ischemia-reperfusion, and sepsis. Inflammatory mediators can disrupt the pulmonary capillary barrier, leading to the influx of a protein-rich edema (58) with severe consequences for gas exchange and the functional integrity of remote organ systems.

Excessive infiltration of polymorphonuclear leukocytes (PMNs) into the lungs has been identified as a pivotal event in the early development of ALI. In various experimental models of lung injury, depletion of PMNs markedly attenuated the severity of lung damage (1). In addition, pulmonary function in patients suffering from ALI negatively correlates with neutrophil counts in the blood (2). PMN recruitment occurs in a cascade-like sequence of activation, sequestration in pulmonary vessels, and transendothelial and transepithelial migration (40). Each migration step is regulated by distinct molecules (39), and the importance of investigating discrete steps of PMN migration in the lung has been emphasized (7). Although neutrophil infiltration has been recognized as a hallmark of ALI, the molecular mechanisms underlying PMN trafficking in the lung still remains incompletely understood. Current therapeutic options are limited to mechanical ventilation with low tidal volumes and other supportive approaches. A specific pharmacotherapy is lacking (58).

Besides other molecule families (3, 43, 44), extracellular adenosine is an essential mediator of endothelial barrier integrity (20) and leukocyte trafficking to inflammatory sites (21). Adenosine is released in response to various stimuli, including hypoxia, tissue damage, and chronic inflammation (5). Adenosine signals through four subtypes of G protein-coupled adenosine receptors (A1, A2a, A2b, and A3) that are ubiquitously expressed on various hematopoietic and nonhematopoietic cells. Each type of adenosine receptor exhibits a distinct pharmacological and physiological profile (15). Activation of adenosine receptors induces a variety of cell responses through changes in intracellular levels of cAMP, diacylglycerol, and inositol triphosphate. Recent studies identified CD39 and CD73, both rate-limiting enzymes for the generation of extracellular adenosine, as critical mediators in pulmonary inflammation (8, 45).

Whereas adenosine receptors A2a and A2b have been recognized as important mediators in inflammatory models (9, 41), A3 has received less attention so far. A3 is expressed in a broad spectrum of tissues, most abundantly in lung and liver (49). A3 signals through Gi, inhibits adenylate cyclase, and activates phospholipase C (33). In addition, A3 binds to MAPKs (52). On activation, A3 is rapidly phosphorylated and desensitized (33). Physiological functions of A3 signaling include mediator release from mast cells and preconditioning (14). The role of A3 in inflammation has been controversial. A3 is involved in directing PMNs to sites of inflammation. Exposure to chemotactic signals leads to upregulation of A3 and translocation from cytosolic granules to the cell surface near the leading edge (6). In some studies, cells that lack A3 lose their ability to effectively migrate toward a chemotactic gradient in vitro and in vivo (6, 27). In contrast, others have found that not inhibition but activation of A3 diminished migration of leukocytes to sites of inflammation (54). In a sepsis model, A3−/− mice exhibited impaired renal and liver function and had a worse 7-day survival compared with wild-type mice. Treatment with an A3 agonist was sufficient to improve organ function and survival in wild-type mice (30). In the lung, activation of A3 protects from ischemia-reperfusion-induced tissue damage (36, 46). In addition, A3 has been implicated in suppressing LPS-induced release of chemotactic cytokines (32, 35), a key event in the early phase of lung inflammation.

We therefore hypothesized that activation of A3 would reduce PMN infiltration in the lungs by inhibiting the release of chemotactic cytokines into the alveolar space. In a murine model of LPS-induced lung injury, we determined the role of A3 for the different migration steps in the lung. Our experimental approach involved pharmacological activation of A3 and the use of A3 gene-deficient (A3−/−) and chimeric mice. We further determined the effects of A3 in the integrity of the pulmonary endothelium and identified the target cell of a specific A3 agonist by means of in vitro transmigration assays.



We used 8- to 12-wk-old, male C57BL/6 mice (20–25 g) in all experiments. Animals were purchased from Charles River Laboratories (Sulzfeld, Germany). A3−/− mice on a C57BL/6 background were obtained from Marlene A. Jacobson (Merck Research Laboratories) (50). All animal protocols were approved by the Animal Care and Use Committee of the University of Tübingen.

Blood counts.

To reveal possible differences between wild-type and A3−/− mice, blood samples were withdrawn from the tail vein, and baseline differential blood cell counts were analyzed (Diff-Quik; Dade Behring, Newark, DE).

Generation of chimeric mice.

To characterize the contribution of A3 on nonhematopoietic (e.g., endothelial and epithelial cells) and hematopoietic cells (e.g., PMNs), chimeric mice were generated by transferring bone marrow between wild-type and A3−/− mice as described earlier (13). Briefly, recipient mice received a lethal irradiation of two doses of 600 rad separated by 4 h. Donor mice were euthanized, femurs and tibias were removed, marrow cavity was flushed with HBSS, and bone marrow cells were collected. Subsequently, 5 × 106 cells in a final volume of 0.3 ml were reinfused into recipient mice immediately on the second irradiation. Bone marrow transference was performed in two directions: bone marrow from A3−/− into wild-type mice (chimeric mice express A3 on nonhematopoietic cells only) and bone marrow from wild-type into A3−/− mice (chimeric mice express A3 on hematopoietic cells only). To confirm efficacy of radiation, sentinel mice did not receive bone marrow. This regimen leads to an almost complete reconstitution (>95%) of mice with donor-derived blood after 6 wk (40). In addition, we transplanted A3−/− bone marrow into A3−/− mice to account for potential effects of the radiation procedure. Chimeric mice were housed in individually ventilated cages for at least 6 wk before experiments, and antibiotics (5 mM sulfamethoxazole and 0.86 mM trimethoprim) were added to drinking water alternating semiweekly.

Murine model of ALI.

To induce an acute pulmonary inflammation, mice were exposed to aerosolized LPS in a custom-built chamber. LPS from Salmonella enteritidis (Sigma-Aldrich, Munich, Germany) was dissolved in sterile saline (500 μg/ml) and aerosolized by an air nebulizer that was connected to the chamber. Mice were allowed to inhale LPS for 30 min. As previously shown (40, 43), LPS inhalation induces pulmonary inflammation with several characteristics of ALI, including PMN trafficking into all compartments of the lung, increase in microvascular leakage, release of chemotactic cytokines, and lung tissue damage (42).

A3 mRNA expression.

Real-time RT-PCR was performed to determine LPS-induced expression of A3 mRNA in whole lung tissue. Wild-type mice inhaled LPS aerosol, and lungs were harvested at indicated times. Control animals did not inhale LPS. Lungs were perfused free of blood, excised, and stored in RNAlater (Ambion) at −80°C. Total RNA was isolated from lung homogenates, and reverse transcription was performed using SuperScript III Reverse Transcriptase Kit (Invitrogen) and oligo(dT) primers. cDNA samples were analyzed with primers for A3 (5′-GCTGCCATCGGGCTCTGTG-3′ and 5′-GCAGGCATAGAAGTGCATCT-3′) on an iCycler iQ Real-Time Detection System (Bio-Rad). Values were determined using the iCycler iQ Real-Time Detection System Software version 3.1 (Bio-Rad). Murine β-actin mRNA (primers 5′-ACATTGGCATGGCTTTGTTT-3′ and 5′-GTTTGCTCCAACCAACTGCT-3′) was amplified in identical reactions to control for the amount of starting template.

Pharmacological activation of A3.

The selective A3 agonist, 2-chloro-N6-(3-iodobenzyl)-adenosine-5′-N-methyluronamide (Cl-IB-MECA; Tocris Bioscience, Bristol, United Kingdom) was used at indicated doses to activate A3 in vitro and in vivo (28). In all in vivo experiments, Cl-IB-MECA was injected intraperitoneally 30 min before LPS exposure. Control groups received DMSO at a similar concentration.

In separate experiments, mice received either the A3 antagonist MRS 1191 (1 mg/kg body wt; Sigma-Aldrich) or the A2a antagonist ZM 241385 (1 mg/kg body wt; Tocris Bioscience, Bristol, United Kingdom) before treatment with Cl-IB-MECA to reveal potential unspecific effects of Cl-IB-MECA on the A2a receptor.

PMN trafficking in the lung.

A flow cytometry-based method was used to determine PMN migration into the different compartments of the lung (pulmonary microvasculature, interstitium, and alveolar air space) (40). Briefly, 24 h after LPS inhalation, an Alexa 633-labeled antibody (Gr-1, clone RB6-8C5) (23) was injected intravenously to label intravascular PMNs. After 5 min, mice were anesthetized by an intraperitoneal injection of ketamine (125 mg/kg; Ratiopharm), xylazine (12.5 mg/kg; Bayer, Leverkusen, Germany), and atropine sulfate (0.025 mg/kg; B. Braun Melsungen), and chests were opened. Lungs were perfused through the spontaneous beating right ventricle with PBS to remove nonadherent leukocytes from the pulmonary vasculature. The trachea was cannulated with a 22-gauge needle, and bronchoalveolar lavage (BAL) fluid was collected. Subsequently, lungs were removed and prepared in the presence of excess unlabeled anti-Gr-1 antibody to prevent possible binding of the injected antibody to extravascular PMNs. Lungs were minced and incubated with digesting enzymes for 1 h. The resulting cell suspension was passed through a 70-μm cell strainer (BD Falcon, Bedford, MA) to separate leukocytes from lung tissue. The cell suspension was incubated with fluorescent antibodies to CD45 (clone 30-F11) and 7/4 (clone 7/4) (24). BAL was stained with CD45, 7/4, and Gr-1. Total cells in BAL and lung were counted, and percentage of PMNs was determined by flow cytometry. In the BAL, PMNs were identified by their typical appearance in forward/sideward scatter and their expression of 7/4 and Gr-1. In the lung, the expression of Gr-1 was used to distinguish intravascular (CD45+, 7/4+, Gr-1+) from interstitial (CD45+, 7/4+, Gr-1) PMNs that were not reached by the injected antibody (40).

Differential blood counts.

To reveal a potential effect of Cl-IB-MECA on the release of PMNs from the bone marrow that might have interfered with the analyses of our migration studies, we performed differential blood counts in mice that were treated with Cl-IB-MECA.


Wild-type and A3−/− mice were anesthetized 24 h after LPS inhalation with or without pretreatment with Cl-IB-MECA. Control mice did not receive LPS. The pulmonary circulation was perfused free of blood, the trachea was cannulated, and the lung was inflated with 4% paraformaldehyde (PFA) for 10 min at 25 cmH2O. The lungs were subsequently removed and fixed in PFA for 24 h. Paraffin-embedded sections (5 μm) were stained for PMNs using the avidin-biotin technique (Vector Laboratories) as previously described (4). Briefly, lung sections were deparaffinized and rehydrated, and nonspecific binding was blocked by incubation with avidin, 10% rabbit serum, and 0.5% fish skin gelatin oil (FSGO). Sections were washed with PBS and incubated overnight with a specific antibody to mouse neutrophils (clone 7/4; Caltag Laboratories) (24). Biotinylated rabbit anti-rat IgG (5 g/ml; Vector Laboratories) was added, incubated for 1 h followed by avidin-biotin-peroxidase complexes (Vectastain Elite ABC Kit; Vector Laboratories), washed with PBS, incubated with diaminobenzidine (DAB kit; Vector Laboratories), and counterstained with hematoxylin.

Chemokine release measurements.

Release of CXCL1 [keratinocyte-derived chemokine (KC)], CXCL2/3 [macrophage inflammatory protein-2 (MIP-2)], TNF-α, and IL-6 into the BAL of wild-type and A1−/− mice was determined by enzyme-linked immunosorbent assay kits (R&D Systems, Minneapolis, MN) 3 h on LPS inhalation. Some animals were pretreated with Cl-IB-MECA 30 min before LPS exposure. Negative controls received no LPS.

A3-dependent pulmonary microvascular leakage.

Microvascular leakage was assessed 6 h after LPS inhalation by the Evans blue dye extravasation technique (19). Evans blue (20 mg/kg; Sigma-Aldrich) was injected into the tail vein 30 min before thoracotomy. The pulmonary vasculature was perfused with PBS through the beating right ventricle to remove intravascular dye. Lungs were excised and homogenized, and Evans blue was extracted as described previously (38). The absorption of Evans blue was determined spectrophotometrically at 620 nm and corrected for the presence of heme pigments: A620 (corrected) = A620 − (1.426 × A750 + 0.030). The amount of Evans blue in lung homogenates was calculated against a standard curve (micrograms Evans blue dye per lung) (19). Some animals received Cl-IB-MECA (100 μg/kg) 30 min before LPS exposure. Control animals did not received LPS.

Cytoskeletal remodeling.

To evaluate whether A3 is involved in cytoskeletal remodeling of pulmonary endothelial cells, we investigated the distribution of F-actin as described previously (43, 51). Briefly, human lung microvascular endothelial (HMVEC-L) cells were allowed to adhere on gelatin-coated glass coverslips overnight (medium with 10% FBS). Cells were then serum-starved for 6 h and stimulated with LPS (100 ng/ml) in presence or absence of Cl-IB-MECA (100 ng/ml). Untreated cells served as control. At indicated times, cells were washed, fixed (4% PFA), permeabilized (0.1% Triton X-100; Sigma-Aldrich), and stained with FITC-phalloidin and DAPI Nucleic Acid Stain (both Invitrogen). Coverslips were mounted on glass slides, and microscopy was performed on a confocal fluorescence microscope (LSM 510; Zeiss, Göttingen, Germany).

In vitro transmigration.

Chemokine-induced transmigration of human PMNs across a monolayer of lung endothelial (HMVEC-L; Lonza, Cologne, Germany) or A549 lung epithelial cells (American Type Culture Collection, Manassas, VA) was studied to evaluate the effect of Cl-IB-MECA on specific cell types. A549 cells were cultured in complete Ham's F-12 medium (Invitrogen-Gibco, Karlsruhe, Germany), including 10% FCS (PAA Laboratories, Cölbe, Germany), 5% antibiotic solution (Sigma-Aldrich), and 200 mM l-glutamine (Invitrogen-Gibco). Cells were plated on the bottom of polycarbonate filter inserts of a Transwell system (3.0-μm pore size, 6.5-mm diameter; Costar, Cambridge, MA) and grown until confluent. Human PMNs were purified from healthy donors (Percoll gradient; GE Healthcare Bio-Sciences, Uppsala, Sweden).

Epithelial cells, PMNs, or both were incubated with Cl-IB-MECA for 30 min at indicated concentrations (10, 1, and 0.1 ng/ml). PMNs (1 × 106) were plated on the top of filters with or without epithelial cells and allowed to migrate toward CXCL2/3 (200 ng/ml; PeproTech, Hamburg, Germany). After 1 h at 37°C, migrated cells were quantified by determination of MPO in the bottom wells. Negative controls did not receive Cl-IB-MECA or CXCL2/3, respectively.

In separate experiments, HMVEC-L cells were cultured in EGM-2MV medium (Lonza) and used in the same Transwell system with minor modifications: cells were plated on the opposite side (top) of filter inserts.

Statistical analysis.

Statistical analysis was performed with GraphPad Prism (version 5.0; GraphPad, San Diego, CA). Differences between groups were assessed by one-way ANOVA followed by post hoc Tukey test. Studies that compared the effect of two factors were analyzed by two-way ANOVA. Data are presented as means ± SD. P < 0.05 was considered statistically significant.


Blood counts.

Genetically altered mice sometimes exhibit increased leukocyte counts in the peripheral blood that may interfere with PMN trafficking and inflammatory response. To reveal such differences, differential blood cell count was performed in wild-type and A3−/− mice. Analyses revealed similar cell counts of all types of leukocytes (Table 1).

View this table:
Table 1.

Baseline cell counts

LPS induces A3 transcription in the lung.

Real-time RT-PCR was performed to reveal LPS-induced transcription of A3 mRNA in the lungs. LPS exposure induced time-dependent A3 mRNA expression. Three hours after LPS exposure, A3 mRNA expression was ∼6-fold (P < 0.05; Fig. 1).

Fig. 1.

LPS-induced expression of adenosine receptor A3 (A3AR) mRNA in the lungs of wild-type mice. LPS inhalation led to continuous transcription of A3. A3 mRNA levels were normalized to β-actin and presented as means ± SD of n = 4 samples; *P < 0.05 vs. control without LPS.

Cl-IB-MECA attenuates LPS-induced PMN trafficking in the lung.

The effect of Cl-IB-MECA on the migration of PMNs into the different compartments of the lung was analyzed quantitatively. Inhalation of LPS increased the number of PMNs in all lung compartments of wild-type and A3−/− mice. No differences between wild-type and A3−/− mice were observed. Pretreatment of wild-type mice with Cl-IB-MECA significantly attenuated PMN accumulation in all compartments (in the vasculature: 0.2 ± 0.04 × 106 vs. 1.0 ± 0.3 × 106; P < 0.05; in the lung interstitium: 0.9 ± 0.3 × 106 vs. 2.9 ± 0.9 × 106; P < 0.05; in the bronchoalveolar space: 0.9 ± 0.1 × 106 vs. 2.2 ± 0.1 × 106; P < 0.05; Fig. 2). Cl-IB-MECA did not reduce migration of PMNs in A3−/− mice. To identify a potential interference with the A2a receptor (53), wild-type mice were pretreated with either the A3 antagonist MRS 1191 or the A2a antagonist ZM 241385 before treatment with Cl-IB-MECA, and accumulation of PMNs in the BAL was analyzed. We found that blocking A3 but not A2a abolished the effect of Cl-IB-MECA, supporting the specificity of this compound (Fig. 3).

Fig. 2.

LPS-induced polymorphonuclear leukocyte (PMN) trafficking into all lung compartments of wild-type (WT) and A3−/− mice. Accumulation of PMNs in the vasculature (IV; A), the lung interstitium (IS; B), and the bronchoalveolar space (BAL; C) were analyzed. Data are means ± SD of n = 4 samples. *P < 0.05 vs. negative control without LPS; #P < 0.05 vs. LPS-treated mice without pretreatment with 2-chloro-N6-(3-iodobenzyl)-adenosine-5′-N-methyluronamide (Cl-IB-MECA).

Fig. 3.

To detect potential unspecific effects of Cl-IB-MECA on the A2a receptor, wild-type mice were pretreated with either the A3 antagonist MRS 1191 or the A2a antagonist ZM 241385 before treatment with Cl-IB-MECA. Accumulation of PMNs in the BAL was analyzed 24 h after LPS inhalation. Values are means ± SD of n = 4 experiments; #P < 0.05 vs. LPS-treated mice without Cl-IB-MECA.

Cl-IB-MECA does not increase number of circulating PMNs.

Adenosine plays a major role in the regulation of hematopoiesis. To reveal potential effects of Cl-IB-MECA on the numbers of circulating PMNs that might have interfered with the analyses of our migration studies, differential blood counts were performed at different time points after Cl-IB-MECA treatment. Cl-IB-MECA did not decrease or increase numbers of circulating PMNs, indicating that the effects of Cl-IB-MECA on pulmonary PMN trafficking was not caused by altered numbers of circulating PMNs (Fig. 4).

Fig. 4.

Effect of Cl-IB-MECA on the numbers of PMNs in the blood of wild-type mice. Values are means ± SD of n = 4 mice.

Quantitative fluorescence-activated cell sorter data were illustrated by immunohistochemistry as shown in Fig. 5.

Fig. 5.

AF: PMN infiltration into the lungs shown by immunohistochemistry. Wild-type and A3−/− mice were exposed to LPS. Lung sections were stained with a specific neutrophil marker (7/4; brown cells). Images are representative of n = 4 experiments.

Activation of A3 reduces LPS-induced release of chemotactic cytokines.

The release of chemotactic cytokines into the alveolar air space initiates the recruitment of inflammatory cells into the lungs. To evaluate the effect of A3 on LPS-induced release of cytokines, levels of CXCL1, CXCL2/3, TNF-α, and IL-6 were determined in the BAL of wild-type and A3−/− mice. At baseline (no LPS), cytokine levels were undetectable in both groups. LPS exposure induced a significant release of all four cytokines in wild-type and A3−/− mice (Fig. 6). In A3−/− mice, LPS-induced release of CXCL2/3 and CXCL1 was higher than in wild-type mice (3,900 ± 200 vs. 1,400 ± 400 pg/ml; P < 0.05 and 2,100 ± 300 vs. 1,600 ± 200 pg/ml; P < 0.05). No difference between groups was found for TNF-α and IL-6. Pretreatment with Cl-IB-MECA significantly diminished the release of TNF-α and IL-6 in wild-type mice (900 ± 400 vs. 2,700 ± 600 pg/ml; P < 0.05 and 100 ± 50 vs. 500 ± 100 pg/ml; P < 0.05). Cl-IB-MECA had no effect in A3−/− mice, confirming the specificity of this compound.

Fig. 6.

Effect of Cl-IB-MECA on LPS-induced release of chemokines into the alveolar air space of wild-type and A3−/− mice. Data are means ± SD of n = 4 samples. *P < 0.05 vs. samples without LPS; #P < 0.05 vs. LPS-treated mice without Cl-IB-MECA; §P < 0.05 vs. wild-type mice within the same treatment group.

Effects of A3 on hematopoietic vs. nonhematopoietic cells.

Next, we sought to define the contribution of A3 on hematopoietic vs. nonhematopoietic cells for PMN trafficking by generation of chimeric mice. Consistent with our initial studies in wild-type and A3−/− mice, we did not observe differences in PMN counts between both groups of chimeric mice (Fig. 7). Surprisingly, pretreatment with Cl-IB-MECA did not alter PMN accumulation in any lung compartment. This finding was in contrast to the results of our studies in wild-type mice (Fig. 2) and suggests that in vivo, Cl-IB-MECA requires both cell types to be effective at modulating pulmonary PMN trafficking. A3−/− mice that had been reconstituted with bone marrow from A3−/− mice served as control. PMN trafficking in these mice was not different from A3−/− mice that had not been radiated, confirming previous reports (43) demonstrating that the radiation procedure alone has no effect on PMN migration (data not shown).

Fig. 7.

Effect of Cl-IB-MECA on PMN trafficking in chimeric mice that express A3 on either hematopoietic (Hematop.) or nonhematopoietic (Nonhematop.) cells. Accumulation of PMNs in the IV (A), the lung IS (B), and the BAL (C) are displayed. Some chimeric mice were pretreated with Cl-IB-MECA. Data are means ± SD of n = 4 samples. No statistical differences were observed.

Cl-IB-MECA ameliorates LPS-induced microvascular leakage.

Increased microvascular leakage is one of the hallmarks of acute pulmonary inflammation. We quantified microvascular permeability by the Evans blue extravasation dye technique. Baseline permeability was not different between wild-type and A3−/− mice. LPS inhalation induced a significant extravasation of Evans blue in the lungs of wild-type (200 ± 66 vs. 49 ± 26 μg/g lung; P < 0.05) and A3−/− mice (180 ± 35 vs. 72 ± 52 μg/g lung; P < 0.05) (Fig. 8). Pretreatment with Cl-IB-MECA reduced Evans blue extravasation in wild-type mice completely (45 ± 13 μg/g lung; P < 0.5). In A3−/− mice, Cl-IB-MECA had no effect on microvascular permeability.

Fig. 8.

Impact of Cl-IB-MECA on LPS-induced microvascular permeability in the lungs of wild-type and A3−/− mice. Microvascular protein leakage was assessed by Evans blue extravasation technique 6 h after LPS exposure. Data are means ± SD of n = 4 samples. *P < 0.05 vs. negative control without LPS; #P < 0.05 vs. LPS-treated wild-type mice without Cl-IB-MECA.

Cytoskeletal remodeling.

After we had established a role of A3 for preventing LPS-induced microvascular permeability, we hypothesized that activation of A3 would reduce the formation of stress fibers in pulmonary endothelial cells, thus maintaining endothelial integrity. To evaluate the effects of A3 on cytoskeletal remodeling, we stained F-actin in pulmonary endothelial cells. LPS induced rapid formation of stress fibers that resulted in cell retraction and formation of intercellular gaps (Fig. 9). Pretreatment with Cl-IB-MECA reduced formation of stress fibers substantially. Although formation of stress fibers may not be specific for altered endothelial permeability, it confirms a biological function of A3 on pulmonary endothelium that accompanies increased permeability as assessed by the extravasation of Evans blue.

Fig. 9.

AF: A3-dependent formation of F-actin in human lung microvascular endothelial (HMVEC-L) cells. LPS induced formation of stress fibers at indicated time points. Pretreatment with Cl-IB-MECA substantially reduced formation of F-actin. Images are representative of 3 experiments with similar results.

A3 on PMNs mediates in vitro migration of human PMNs.

To translate the findings of our studies to the human cell system, we evaluated the effect of Cl-IB-MECA on chemokine-induced migration of human PMNs across a monolayer of HMVEC-L or A549 cells. Activation of A3 on PMNs resulted in a significant decrease in transendothelial migration (10 ng/ml: 30,000 ± 1,000 vs. 86,000 ± 2,000; 1 ng/ml: 28,000 ± 6,000 vs. 86,000 ± 2,000; 0.1 ng/ml: 26,000 ± 10,000 vs. 86,000 ± 2,000; all P < 0.05; Fig. 10). Simultaneous activation of A3 on PMNs and endothelial cells did not show an additional effect. Pretreatment of endothelial cells alone did not reduce PM migration.

Fig. 10.

In vitro effect of Cl-IB-MECA on chemokine-induced migration of human PMNs across a layer of HMVEC-L cells. Transmigration was significantly reduced when PMNs were pretreated with Cl-IB-MECA (A). Pretreatment of HMVEC-L cells alone did not alter PMN migration (B). Simultaneous pretreatment of PMNs and endothelial cells did not further reduce migration (C). Data are means ± SD of n = 4 experiments (each in duplicates); *P < 0.05 vs. negative control without CXCL2/3; #P < 0.05 vs. positive control without Cl-IB-MECA.

Similar results were obtained in appropriate chemotaxis experiments with A549 cells. Pharmacological activation of A3 on PMNs reduced transepithelial migration significantly (10 ng/ml: 37,000 ± 1,000 vs. 414,000 ± 24,000; 1 ng/ml: 24,000 ± 1,000 vs. 414,000 ± 24,000; 0.1 ng/ml: 18,000 ± 6,000 vs. 414,000 ± 24,000; all P < 0.05; Fig. 11). Pretreatment of epithelial cells alone did not alter PMN migration. Simultaneous treatment of PMNs and epithelial cells did not show an additional effect.

Fig. 11.

The effect of Cl-IB-MECA on chemokine-induced transmigration across A549 cells was evaluated in vitro. Transmigration was significantly reduced when PMNs were pretreated with Cl-IB-MECA (A). Pretreatment of epithelial cells alone did not alter PMN migration (B). Simultaneous pretreatment of PMNs and A549 cells did not further reduce migration (C). Data are means ± SD of n = 4 experiments (each in duplicates); *P < 0.05 vs. negative control without CXCL2/3; #P < 0.05 vs. positive control without Cl-IB-MECA.


This study was designed to characterize the role of A3 in a murine model of LPS-induced lung injury. We found that A3 is critically involved in mediating pulmonary PMN trafficking. Activation of A3 with the specific agonist Cl-IB-MECA decreased PMN migration into all lung compartments. Reduced accumulation of PMN was associated with decreased release of relevant cytokines into the alveolar air space. In addition, endothelial integrity was maintained in mice that were pretreated with Cl-IB-MECA.

Extracellular adenosine is a potent modulator in various inflammatory pathways. Adenosine is generated by a cascade-like enzymatic process involving the ecto-apyrase (CD39, conversion of ATP/ADP to AMP) and the ecto-5′-nucleotidase (CD73, conversion of AMP to adenosine). Both rate-limiting enzymes have been previously implicated in attenuating ALI and inflammation in models of ambient hypoxia (10, 11), cyclic mechanical stretch (8), and LPS-induced lung injury (45).

Research on the significance of A3 in inflammatory disorders has produced conflicting results. Inoue et al. (27) found that in a murine sepsis model, migration of leukocytes into peritoneum and lungs was reduced in A3−/− mice. Moreover, A3−/− mice survived longer than wild-type mice (27), leading the authors to suggest a proinflammatory effect of A3 in this model. In contrast, Lee and coworkers (30) found improved survival and organ function in animals that were pretreated with the A3 agonist Cl-IB-MECA. Both groups used the same disease model [cecal ligation and puncture (CLP)], however, the severity of sepsis might have been different at best. Although mechanisms underlying the dual role of A3 in regulating inflammatory pathways have not been elucidated completely, differences in disease model, method of treatment, and targeted cell type have been emphasized as important parameters that account for the “two personalities” of A3 (12, 56). A key role of A3 has been demonstrated in numerous models of ischemia-reperfusion-induced injury in various organ systems, including brain (57), kidney (31), and heart (34). Although a proinflammatory role of A3 was supported by some (31), other studies implied A3 as a rather protective mediator in inflammatory disorders. Treatment with selective A3 agonists effectively reduced leukocyte infiltration and infarct size after myocardial ischemia (17, 34) and reduced mortality in experimental sepsis models (22, 30).

Evidence for a significant involvement of A3 in pulmonary inflammation originates from studies demonstrating that A3 is highly expressed in the lung (49) and mRNA transcription in lung tissue is upregulated on inflammatory stimuli (37). In the present study, LPS inhalation also increased A3 mRNA in lung homogenates. Pharmacological activation of A3 has been widely used to demonstrate protective effects of A3 in pulmonary inflammation. In an in vivo model of ischemia-reperfusion-induced lung injury, pretreatment with selective A3 agonists reduced organ damage, edema formation, apoptosis, and leukocyte infiltration as assessed by myeloperoxidase activity (46). These protective effects were associated with the upregulation of ERK1/2 (36). These results are consistent with the present study supporting a specific role of A3 in regulating PMN trafficking and microvascular permeability.

Early research suggested a distinct role for A3 in mediating chemotaxis of eosinophils (29). More recent studies identified A3 as a key molecule for the regulation of chemotaxis of PMNs. The release of chemotactic cytokines into the alveolar air space is a key event in the early phase of lung inflammation, initiating the recruitment of leukocytes to the lung. Chemokines are produced and released by alveolar macrophages and type II cells (55). In the present study, activation of A3 reduced the accumulation of TNF-α and IL-6 in the BAL (Fig. 4). This extends previous evidence demonstrating that activation of A3 reduces the release of TNF-α from human macrophages (48) and into the plasma of wild-type mice (50) and might represent one major mechanism that mediated A3-dependent leukocyte recruitment in the present study.

It appears important to recognize that concentrations of chemokines vary between the different compartments (e.g., plasma and BAL). In this context, our model substantially differs from others (e.g., sepsis models) in that with LPS inhalation, alveolar macrophages and type II cells are directly targeted and release of chemokines is initiated in the alveolar space. LPS-induced accumulation of CXCL1 was significantly higher in A3−/− compared with wild-type mice. However, this did not translate into increased migration of PMNs in A3−/− compared with wild-type mice. Insufficient activation of A3 might have prevented a detectable protective response in wild-type mice. This hypothesis is supported by studies showing that effects of A3 antagonists require activation of A3, e.g., by genetic removal of adenosine deaminase (ADA) (59, 60).

Adenosine plays a major role in the regulation of hematopoiesis. Effects of specific adenosine receptors on the release of bone marrow-derived cells into the circulation may represent an important mechanism of adenosine-dependent modulation of inflammatory responses that may interfere with the analyses of leukocyte trafficking. Recent research provided evidence that Cl-IB-MECA alone did not increase numbers of circulating leukocytes in vivo. However, Cl-IB-MECA can potentiate in vitro effects of some hematopoietic growth factors (26) and has been suggested to stimulate granulopoiesis under conditions of myelosuppression (e.g., pretreatment with cytotoxic agents such as 5-FU) (25). In most of these studies, multiple doses of Cl-IB-MECA were required to induce granulopoiesis. In our model, Cl-IB-MECA did not affect numbers of circulating PMNs.

We used chimeric mice to characterize the contribution of A3 on hematopoietic vs. nonhematopoietic cells. Evidence from recent in and ex vivo work (18, 54) and from our in vitro transmigration assays with human PMNs (Figs. 8 and 9) implied a distinct role of A3 on PMNs. In a model of myocardial ischemia-reperfusion, pharmacological activation of A3 did not reduce the number of infiltrating PMNs when A3 had been removed from hematopoietic cells (18). In our model, Cl-IB-MECA lost its effect in our model when A3 had been removed from either compartment (Fig. 5), indicating that in vivo, A3 on hematopoietic and nonhematopoietic cells is required. However, the experimental setup between both studies differed fundamentally. Regulation of leukocyte trafficking in the lung with its unique architecture (capillary size, compartments, presence of alveolar macrophages, etc.) may not reflect conditions in other organs such as the heart. Moreover, in the study by Ge et al. (18), PMN accumulation was assessed semiquantitatively by light microscopy and did not distinguish between adherent and transmigrated cells. In addition, reversal bone marrow transplantation was not performed in this study, leaving the role of A3 on nonhematopoietic cells unaddressed.

Disruption of endothelial integrity is another key event during acute lung inflammation. In the present study, pharmacological activation of A3 prevented LPS-induced increase in microvascular permeability almost completely. In human pulmonary microvascular endothelial cells, LPS induced cell remodeling. Although activation of A3 on endothelial or epithelial cells was not sufficient to affect PMN migration in vitro, formation of stress fibers was substantially reduced when endothelial cells were pretreated with Cl-IB-MECA, indicating a significant role for A3 in mediating microvascular permeability and supporting the theory of distinct regulation and molecular requirements of microvascular permeability and leukocyte trafficking (16).

In summary, we have identified A3 as a critical mediator in LPS-induced inflammation of the lung. In particular, A3 controls PMN trafficking into all lung compartments and maintains endothelial integrity. Pharmacological activation of A3 may represent a promising approach to limit LPS-induced inflammatory response in the lung.


This study was supported by the German Research Foundation (Grant RE 1683/3-1 to J. Reutershan).


No conflicts of interest, financial or otherwise, are declared by the author(s).


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
  60. 60.
View Abstract