Inosine Exerts a Broad Range of Antiinflammatory Effects in a Murine Model of Acute Lung Injury

Abstract

Objective

To investigate the effects of inosine on the acute lung inflammation induced by lipopolysaccharide (LPS) in vivo and on the activation and cytotoxicity elicited by proinflammatory cytokines on human lung epithelial (A549) cells in vitro.

Summary Background Data

Inosine is an endogenous purine recently shown to exert immunomodulatory and antiinflammatory effects.

Methods

Mice challenged with intratracheal LPS (50 μg) were treated after 1, 6, and 12 hours with inosine (200 mg/kg intraperitoneal) or vehicle. After 24 hours, bronchoalveolar lavage fluid was obtained to measure proinflammatory (tumor necrosis factor-alpha [TNF-α], interleukin [IL]-1β, IL-6), and antiinflammatory (IL-10, IL-4) cytokines, chemokines (MIP-1α and MIP-2), myeloperoxidase activity and total cell counts, nitric oxide production, and proteins. Lung histology and immunohistochemical detection of 3-nitrotyrosine, a marker of nitrosative stress, were performed in inflated-fixed lungs. In vitro, cell viability and production of the chemokine IL-8 were evaluated in A549 cells stimulated with a mixture of cytokines in the presence or absence of inosine.

Results

Inosine downregulated the LPS-induced expression of TNF-α, IL-1β, IL-6 and MIP-2 and tended to reduce MIP-1α, whereas it enhanced the production of IL-4. Total leukocyte counts, myeloperoxidase, nitric oxide production, and proteins were all significantly decreased by inosine. The purine also improved lung morphology and suppressed 3-nitrotyrosine staining in the lungs after LPS. Inosine attenuated the cytotoxicity and the expression of IL-8 induced by proinflammatory cytokines in A549 cells.

Conclusions

Inosine largely suppressed LPS-induced lung inflammation in vivo and reduced the toxicity of cytokines in lung cells in vitro. These data support the proposal that inosine might represent a useful adjunct in the therapy of acute respiratory distress syndrome.

Acute respiratory distress syndrome (ARDS) is a major complication of conditions such as sepsis, trauma, and severe pneumonia. Despite significant advances in our understanding of its underlying pathogenic mechanisms, no satisfying therapy has emerged so far, and treatment of ARDS remains largely supportive. ^1,2 ARDS is characterized by an overwhelming lung inflammation involving the local recruitment and activation of polymorphonuclear neutrophils ^2–5 and the release of proinflammatory mediators, ^6,7 proteases, and both reactive oxygen and nitrogen species. ^8,9 Eventually, this results in alveolocapillary damage, high permeability pulmonary edema, altered lung mechanics, and severe gas exchange abnormalities. ^1,2 Although ARDS develops in particular clinical settings not easily reproducible experimentally, several animal models give useful information about its pathophysiologic mechanisms. In particular, the in vivo intratracheal administration of lipopolysaccharide (LPS), a component of the wall of gram-negative bacteria, has gained wide acceptance as a clinically relevant model of severe lung inflammation. ^10–12

Inosine is a naturally occurring purine formed from the breakdown of adenosine by adenosine deaminase. ¹³ It has been shown recently that inosine can participate in receptor-mediated signaling, ^14,15 and several lines of investigations support an important role for inosine in modifying the inflammatory response. ^14–17 We have recently reported that in vitro inosine reduced the production of proinflammatory cytokines by murine macrophages stimulated by LPS, ¹⁶ whereas in vivo it modulated the expression of pro- and antiinflammatory cytokines ¹⁶ and provided protective effects on both intestinal and vascular function in murine models of endotoxic shock. ¹⁸ Although the mechanisms underlying the beneficial effects of inosine remain incompletely understood, the results of the aforementioned studies support the concept that inosine might represent an interesting immunomodulator agent in conditions associated with severe systemic inflammation. However, there is no information regarding the influence of inosine treatment on more compartmentalized models of inflammation. Further, although the effects of inosine in vitro have been partially characterized on rodent cell lines, ^14–16,19 limited information is currently available regarding its actions on human cell lines. ²⁰ Therefore, the present study was undertaken to evaluate the effects of inosine on acute lung inflammation induced by the intratracheal instillation of LPS in vivo. In addition, the effects of inosine were evaluated in an in vitro model using the human lung epithelial cell line A549.

METHODS

In Vitro Studies

Cell Culture Conditions

Human alveolar epithelial cells (A549, American Tissue Culture Collection, Manassas, VA) were grown in RPMI 1640 medium supplemented with 10% fetal calf serum (FCS) (Life Technologies, Frederick, MD). Cells were passaged every 4 days, removed by trypsin (0.05%) EDTA (0.02%) and then cultured (5% CO₂, 37°C) to confluence in 96- or 24-well plates. For the experiments, the cells were incubated in RPMI supplemented with 0.5% FCS instead of 10%. Immunostimulation of the cells was achieved by a mixture of cytokines consisting of human tumor necrosis factor-alpha (TNF-α; 40 ng/mL), human interleukin (IL)-1β (10 ng/mL), and human interferon-gamma (IFN-γ; 100 ng/mL). The concentrations of the cytokines were determined on the basis of pilot experiments as well as from previously published data. ²¹ Cells were stimulated for 24 hours, in the absence or presence of inosine, added to the cells 1 hour before stimulation.

Measurement of Cell Viability

Cell viability was assessed by measuring the mitochondrial-dependent reduction of MTT (3–4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) to formazan. ²² Cells grown in 96-well plates were stimulated for 24 hours, after which the medium was removed and the cells incubated for 60 minutes with MTT (100 μL, 0.5 mg/mL in culture medium), followed by aspiration and solubilization of the cells in 100 μL DMSO. The amount of formazan formed was quantified by measuring the absorbance of the solution at 540 nm. The concentrations of inosine used in these experiments ranged from 100 μmol/L to 10 mmol/L.

Measurement of Interleukin-8 Expression

Cells were grown to confluence in 24-well plates. To ensure maximal stimulation, cells were then incubated for 24 hours with TNF-α (40 ng/mL), IL-1β (10 ng/mL), and IFN-γ (100 ng/mL) in the presence or absence of inosine, added to the wells 1 hour before stimulation. A concentration of 1 mmol/L inosine was used in these experiments because we previously found that 1 mmol/L inosine allowed a maximal inhibition of cytokine production by immunostimulated macrophages. ¹⁶ In addition, previous data from the literature have indicated that the interstitial levels of inosine increase to greater than 1 mmol/L in inflammatory conditions and shock, resulting from enhanced deamination of adenosine. ^23,24 After 24 hours, the cell medium was aspirated and centrifuged to remove any cell debris, and the concentration of IL-8 was assessed by enzyme-linked immunosorbent assay (ELISA; R&D Systems, Minneapolis, MN) in 4 to 5 wells/condition.

Studies Performed In Vivo

In vivo studies were performed in accordance with National Institutes of Health guidelines and with the approval of the local institutional animal care and use committee.

Intratracheal Lipopolysaccharide Administration

Male BALB/c mice, 8 to 10 weeks old, were anesthetized with a mixture of ketamine (80 mg/kg) and xylazine (30 mg/kg) given intraperitoneally. A 1-cm midline cervical incision was made, exposing the trachea. Intratracheal administration of LPS (Escherichia coli, O127:B8) or vehicle (phosphate-buffered saline [PBS], pH 7.4), was performed with a bent 27G tuberculin syringe (in a volume of 100 μL), as described previously. ^10–12 The cervical incision was closed with 5-0 silk suture and the mice were returned to their cage. The animals recovered rapidly after surgery.

Bronchoalveolar Lavage

Twenty-four hours after the surgery, the mice were reanesthetized with ketamine and xylazine given intraperitoneally. The animals were bled by transsection of the inferior vena cava to reduce hemorrhage into the lungs. Bronchoalveolar lavage was performed by the intratracheal instillation of 1 mL PBS (pH 7.4) into the exposed lungs (maintained within the thoracic cavity). The lavage fluid was infused a total of three times into the lungs before final collection. The bronchoalveolar lavage fluid (BALF) was then centrifuged at 5,000 rpm for 10 minutes and the cell-free supernatant was frozen at −70°C until further analysis. The volume of cell-free supernatant was measured for each animal. The cells were resuspended in a volume of 0.5 mL (0.4 mL PBS and 0.1 mL 0.4% Trypan blue) and total cell counts were performed with a hemocytometer.

Treatment Groups

Animals challenged with intratracheal LPS were treated with inosine, 200 mg/kg (n = 14), or vehicle (isotonic saline, 0.5 mL, n = 14, control group), administered intraperitoneally, 1 hour after surgery. The treatment was repeated after 6 and 12 hours. In addition, a group of eight mice (sham group) was challenged with intratracheal PBS instead of LPS.

Measurements

Protein Assay

The amount of proteins in the BALF was assayed using the Bradford assay. Proteins are expressed in mg protein/mL BALF.

Myeloperoxidase Activity

The activity of myeloperoxidase, an indicator of neutrophil accumulation, was directly measured in the BALF. An aliquot of BALF (20 μL) was mixed with 1.6 mmol/L tetra-methyl-benzidine and 1 mmol/L hydrogen peroxide. Myeloperoxidase activity was then measured as the change in absorbance at 650 nm at 37°C using a Spectramax microplate reader (Molecular Devices, Sunnyvale, CA). Results are expressed as milliunits of myeloperoxidase activity/mL BALF.

Nitrate/Nitrite Concentration

The pulmonary production of nitric oxide was determined by the measurement of nitrate and nitrite, the stable end products of nitric oxide metabolism, in the BALF. First, nitrate was reduced to nitrite by incubation with nitrate reductase (610 mU/mL) and NADPH (170 mmol/L) at room temperature for 3 hours. After 3 hours, the nitrite concentration in the samples was measured by the Griess reaction, by adding 100 μL Griess reagent (0.1% naphthalethylenediamine-dihydrochloride in H₂O and 1% sulfanilamide in 5% concentrated H₃PO₄; vol 1:1). The optical density at 550 nm (OD 550, corrected for absorbance at 650 nm) was measured in a Spectramax microplate reader. Nitrite concentrations were calculated by comparison of OD 550 of standard solutions of sodium nitrite prepared in PBS. Results are expressed as nmol of nitrate and nitrite/BALF.

Cytokines and Chemokines

Amounts of the proinflammatory cytokines TNF-α, IL-1β, and IL-6, the antiinflammatory cytokines IL-10 and IL-4, and the chemokines MIP-1α (a CC chemokine) and MIP-2 (a CXC chemokine) were determined in the BALF using commercially available ELISAs in accordance with the protocol provided by the manufacturer (R&D Systems). For the measurement of proinflammatory cytokines and chemokines in mice challenged with LPS, BALF was diluted 1:2 to 1:5, whereas it was assayed undiluted for IL-4 and IL-10. BALF was assayed undiluted in sham animals.

Lung Histology and Immunohistochemistry

Histopathologic changes induced by LPS were evaluated in five mice treated with inosine and five mice treated with vehicle. Twenty-four hours after surgery, the animals were anesthetized and killed by exsanguination, and the lungs were inflated-fixed with 4% paraformaldehyde. Paraffin-embedded lungs were sectioned at 3 μm and stained with hematoxylin and eosin for morphologic analysis, or processed for the immunohistochemical determination of 3-nitrotyrosine (3-NT), as previously described. ²⁵ 3-NT is a marker of nitrosative stress, and more specifically of the generation of peroxynitrite, a powerful oxidant species formed from the reaction of nitric oxide with the superoxide radical. ²⁶ Sections were deparaffinized in xylene and rehydrated in decreasing concentrations (100%, 95%, and 70%) of ethanol followed by a 10-minute incubation in PBS (pH 7.4). Sections were treated with 0.3% hydrogen peroxide for 15 minutes to block endogenous peroxidase activity and then rinsed briefly in PBS. Nonspecific binding was blocked by incubating the slides for 1 hour in PBS containing 2% goat serum. Rabbit polyclonal antinitrotyrosine antibody (Upstate Biotechnology, Lake Placid, NY) was applied in a dilution of 1:200 for 2 hours at room temperature. After extensive washing (5 × 5 minutes) with PBS, immunoreactivity was detected with a biotinylated goat antirabbit secondary antibody and the avidin-biotin-peroxidase complex (ABC), both supplied in the Vector Elite kit (Vector Laboratories, Burlingame, CA). Color was developed using Ni-DAB substrate (95 mg diaminobenzidine, 1.6 g NaCl, 2 g nickel sulfate in 200 mL 0.1 mol/L acetate buffer). Sections were then counterstained with nuclear fast red, dehydrated, and mounted in Permount. Photomicrographs were taken with a Zeiss Axiolab microscope equipped with a Fuji HC-300C digital camera (Fuji Medical Systems, Inc., Stamford, CT).

Statistical Analysis

All the data were statistically evaluated using analysis of variance. When the relevant F values were significant at the 5% level, further pairwise comparisons were made using the Tukey posthoc test. Statistical significance was assigned to P < .05.

Chemicals

Unless otherwise specified, all chemicals were purchased from Sigma Chemicals (St. Louis, MO).

RESULTS

Inosine Reduces the Cytotoxicity and the Production of Interleukin-8 Elicited by Proinflammatory Cytokines on Lung Epithelial Cells

In these in vitro experiments, we did not use LPS as a stimulating agent because we found in pilot experiments that the cytotoxicity of LPS, even at doses as high as 100 μg/mL, was only marginal in A549 cells. Also, LPS stimulation produced only a weak increase in the expression of IL-8 by these cells. The cells were therefore stimulated by a mixture of the proinflammatory cytokines TNF-α, IL-1β, and IFN-γ. As illustrated in Figure 1A, this stimulation resulted in a significant decrease in cell viability, an effect that was partially, but significantly improved by inosine at concentrations of 1 and 10 mmol/L. A further effect of cytokine stimulation was the production of large amounts of IL-8 by A549 cells. In unstimulated cells, IL-8 concentration was 51.9 ± 28.1 pg/mL, increasing to 6,955 ± 35 pg/mL in stimulated cells (see Fig. 1B). In the presence of 1 mmol/L inosine, the increase in IL-8 was significantly reduced to 4,910 ± 59 pg/mL.

Figure 1. Effect of inosine on the toxicity and the expression of interleukin (IL)-8 elicited by proinflammatory cytokines in lung epithelial cells. (A) Confluent human alveolar epithelial cells (A549) were stimulated in 96-well plates with a mixture of cytokines comprising tumor necrosis factor-alpha (40 ng/mL), IL-1β (10 ng/mL), and interferon-gamma (100 ng/mL). Cell viability was assessed after 24 hours by the mitochondrial-dependent reduction of MTT to formazan, in the presence or absence of inosine (0.1–10 mmol/L), added to the wells 1 hour before the cytokines. Cell viability (n = 6 wells/condition) is expressed as a percentage of control (C). (B) A549 cells grown in 24-well plates were stimulated with cytokines as described above, in the absence or presence of inosine (1 mmol/L). The concentration of IL-8 was determined in the culture medium after 24 hours (n = 4 or 5 wells/condition). Means ± SEM. #P < .05 vs. control (unstimulated) conditions. *P < .05; significant reduction of the effects of cytokines in the presence of inosine.

Inosine Reduces Lung Leukocyte Accumulation After Intratracheal Instillation of Lipopolysaccharide

A small amount of leukocytes were recovered in the BALF from sham mice (2.07 ± 0.13 × 10⁴ cells/BALF, Fig. 2A), and this was associated with a very low myeloperoxidase activity (5.5 ± 0.6 mU/mL BALF, Fig. 2B). Administration of LPS elicited a massive recruitment of leukocytes (76.7 ± 12.9 × 10⁴ cells/BALF) as well as a marked increase in myeloperoxidase activity (149.9 ± 24.6 mU/mL BALF), indicating the presence of a significant proportion of polymorphonuclear cells. Inosine treatment largely suppressed the accumulation of leukocytes in the alveolar spaces, as indicated both by a significant reduction in cell counts (19.5 ± 4.9 × 10⁴ cells/BALF) and myeloperoxidase activity (74.6 ± 13.3 mU/mL BALF).

Figure 2. Total leukocyte counts and myeloperoxidase activity in the bronchoalveolar lavage fluid (BALF). BALB/c mice received an intratracheal instillation of lipopolysaccharide (LPS; 50 μg) and were treated with inosine (n = 14) or vehicle (n = 14). Sham mice received vehicle instead of LPS intratracheally (n = 8). Total leukocyte counts (A) and myeloperoxidase activity (B) were determined in BALF obtained 24 hours after LPS. Means ± SEM. †P < .05 vs. sham. *P < .05 inosine vs. vehicle-treated LPS mice.

Inosine Reduces the Expression of Chemokines After Lipopolysaccharide Administration

The levels of both MIP-1α and MIP-2 were undetectable in sham animals (Fig. 3). After LPS administration, a large increase in MIP-1α (649.8 ± 78.2 pg/mL BALF) and MIP-2 (434.8 ± 42 pg/mL BALF) was noted in the BALF. Inosine treatment significantly reduced the level of MIP-2 to 298.9 ± 24.9 pg/mL BALF, whereas it marginally reduced (P = .06, t test vs. vehicle-treated mice) the level of MIP-1α to 445.1 ± 64.1 pg/mL BALF.

Figure 3. Concentrations of the chemokines MIP-1α and MIP-2 in the bronchoalveolar lavage fluid (BALF). Mice were challenged at baseline with lipopolysaccharide (LPS; 50 μg intratracheally) or vehicle (sham group, n = 8). Mice receiving LPS were treated with inosine (n = 14) or vehicle (n = 14). After 24 hours, BALF was obtained and assayed for the concentrations of MIP-1α (A) and MIP-2 (B). Means ± SEM. †P < .05 vs. sham. *P < .05 inosine vs. vehicle-treated LPS mice.

Inosine Downregulates the Expression of TNF-α, IL-1β, and IL-6 and Increases the Production of IL-4 in Lungs Exposed to Lipopolysaccharide

Figure 4 illustrates the results of the measurements of the proinflammatory cytokines TNF-α, IL-1β, and IL-6 in the BALF. These cytokines were not detectable in sham animals, whereas they all markedly increased after LPS administration (values for TNF-α, IL-1β, and IL-6, respectively: 11.69 ± 1.6 ng/mL BALF; 328.3 ± 40.5 pg/mL BALF; and 1,615.6 ± 159.3 pg/mL BALF). Treatment with inosine significantly suppressed the expression of all three cytokines to 7.18 ± 0.84 ng/mL BALF (TNF-α), 195.1 ± 30.2 pg/mL BALF (IL-1β), and 914.9 ± 156.9 pg/mL BALF (IL-6).

Figure 4. Concentrations of tumor necrosis factor-alpha (TNF-α), interleukin (IL)-1β, and IL-6 in the bronchoalveolar lavage fluid (BALF). Cytokines were measured in the BALF obtained 24 hours after an intratracheal instillation of lipopolysaccharide (LPS; 50 μg) or vehicle (sham group, n = 8). Mice receiving LPS were treated with inosine (n = 14) or vehicle (n = 14). LPS induced a massive increase of TNFα (A), IL-1β (B), and IL-6 (C) that was significantly suppressed by inosine. Means ± SEM. †P < .05 vs. sham. *P < .05 inosine vs. vehicle-treated LPS mice.

In contrast to the absence of proinflammatory cytokines in the BALF from sham animals, we found that these animals expressed a basal level of the antiinflammatory cytokines IL-4 and IL-10 (respectively, 47.7 ± 4.5 pg/mL BALF and 194.9 ± 30.1 pg/mL BALF). Although LPS administration tended to reduce the levels of IL-10, there was no statistically significant difference between groups (LPS–vehicle: 108.54 ± 18.8 pg/mL BALF; LPS–inosine 141.6 ± 27.9 pg/mL BALF). Regarding IL-4, a significant increase, with respect to the value measured in sham animals, was noted in LPS animals receiving inosine (90.2 ± 10.8 pg/mL BALF), whereas no significant change was noted in vehicle-treated animals (69.7 ± 11.4 pg/mL BALF (Fig. 5).

Figure 5. Concentrations of the antiinflammatory cytokines interleukin (IL)-4 and IL-10 in the bronchoalveolar lavage fluid (BALF). Mice received at baseline an intratracheal administration of lipopolysaccharide (LPS; 50 μg) or vehicle (sham group, n = 8). Mice receiving LPS were treated with inosine (n = 14) or vehicle (n = 14). After 24 hours, BALF was obtained and the concentrations of IL-4 (A) and IL-10 (B) were determined by enzyme-linked immunosorbent assay. Compared with sham animals, IL-4 was significantly increased in the BALF of LPS mice treated with inosine, whereas no effect was observed in LPS mice treated with vehicle. In contrast, the levels of IL-10 showed no significant differences between the groups. Means ± SEM. †P < .05 vs. sham.

Inosine Reduces the Formation of High-Permeability Edema and Decreases the Production of Nitric Oxide in Lungs Exposed to Lipopolysaccharide

A hallmark of ARDS is the development of high-permeability edema, characterized by a high protein content of the edema fluid. Such abnormalities were present in the lungs of mice exposed to LPS (Fig. 6A), the protein concentration in the BALF increasing from 0.20 ± 0.02 mg/mL BALF in sham animals to 1.13 ± 0.08 mg/mL BALF in mice challenged with intratracheal LPS. The concentration of protein in the BALF was significantly reduced in LPS mice receiving inosine to a value of 0.77 ± 0.07 mg/mL BALF. A further effect of LPS administration was a significant enhancement of the production of nitric oxide, as indicated by an increase in nitrite/nitrate (see Fig. 6B) in the BALF (from 1.86 ± 0.19 nmol/mL BALF in sham mice to 8.21 ± 0.67 nmol/mL BALF in mice exposed to LPS), an effect that was significantly suppressed by inosine treatment (5.90 ± 0.71 nmol/mL BALF).

Figure 6. Bronchoalveolar lavage fluid (BALF) concentrations of proteins and nitrite/nitrate. BALB/c mice received an intratracheal instillation of lipopolysaccharide (LPS; 50 μg) and were treated with inosine (n = 14) or vehicle (n = 14). Sham mice received vehicle instead of LPS intratracheally (n = 8). After 24 hours, BALF was obtained to measure proteins as an index of high-permeability pulmonary edema (A), as well as nitrite/nitrate (NO_x, B), to evaluate the pulmonary production of nitric oxide. The increase in protein concentration and NO_x induced by LPS was significantly reduced by inosine. Means ± SEM. †P < .05 vs. sham. *P < .05 inosine vs. vehicle-treated LPS mice.

Inosine Reduces Morphologic Damage and the Formation of 3-Nitrotyrosine in Lungs Exposed to Lipopolysaccharide

As illustrated in Figure 7, the lungs of mice exposed to LPS showed marked inflammatory alterations characterized by a thickening of the alveolocapillary membrane, the presence of alveolar hemorrhage, and massive extravasation of both mono- and polymorphonuclear leukocytes into the alveolar spaces (Fig. 7A and Fig. 7B are representative sections from two different animals challenged with LPS). In contrast, the histologic damage was less pronounced in mice treated with inosine, where the amount of both erythrocytes and leukocytes in the alveolar spaces was clearly reduced (see Figs. 7C and 7D).

Figure 7. Lung morphology in mice challenged with lipopolysaccharide (LPS). Representative histologic sections of lungs harvested 24 hours after the intratracheal instillation of LPS in mice treated with vehicle (A, B) or inosine (C, D). The morphologic alterations induced by LPS (infiltration of the alveolar spaces with mono- and polymorphonuclear cells, presence of alveolar hemorrhages and exudates, thickening of the alveolar septa) were reduced by inosine treatment. Pictures are representative of n = 5 mice/treatment group. Magnification ×400.

As shown in Figure 8, a diffuse positive immunostaining for 3-nitrotyrosine was noted in infiltrating leukocytes and in the alveolar epithelial cells of mice challenged with LPS. Although most of the staining was localized in the cytoplasm, some nuclear staining was evident as well. In mice treated with inosine, the staining was not abolished but markedly suppressed, both in leukocytes and in the alveolar walls.

Figure 8. Immunohistochemical localization of 3-nitrotyrosine. Lungs were harvested 24 hours after the intratracheal administration of lipopolysaccharide (LPS), followed by treatment with inosine or vehicle. An intense positive staining for 3-nitrotyrosine was detected in infiltrating leukocytes (A) and alveolar epithelial cells (B) in mice treated with vehicle. In contrast, treatment with inosine markedly suppressed the staining both in inflammatory cells (C) and alveolar epithelial cells (D). Pictures are representative of n = 5 mice/treatment group. Magnification ×1,000.

DISCUSSION

The main finding of this study was that inosine exerted potent antiinflammatory effects in lungs exposed to LPS. Inosine downregulated the expression of proinflammatory cytokines and chemokines, reduced the infiltration of activated polymorphonuclear neutrophils in the airways, decreased pulmonary edema, reduced nitrosative stress, and improved lung morphology. In addition, inosine attenuated the cytokine-induced expression of IL-8 in human alveolar epithelial cells and also exerted direct cytoprotective actions in these cells.

Inosine Downregulates the Expression of Chemokines In Vitro and In Vivo: Effects on Leukocyte Trafficking

The presence of infiltrating leukocytes is the hallmark of pulmonary inflammation associated with acute lung injury. ^3,27 Recruitment and activation of leukocytes at sites of inflammation are orchestrated by a number of distinct events, including the generation of proinflammatory cytokines, the expression of cell-surface adhesion molecules, and the production of chemotactic molecules, most significantly chemokines. ^28,29 Chemokines are classified into α-chemokines (CXC), which mostly attract neutrophils, and β-chemokines (CC), which mostly attract monocytes, T cells, basophils, and eosinophils. ³⁰ A central role of the CXC chemokine IL-8 in the pathogenesis of ARDS has been recently highlighted in several clinical studies showing a direct correlation between the levels of IL-8 in BALF, the severity of lung inflammation, and the death rate. ^31–36 Further, anti-IL-8 antibodies have been shown to exert potent protective effects in experimental models of ARDS. ^37,38 In rodents, where no homolog of IL-8 has been described so far, the α chemokines MIP-2 and KC play the major role in attracting neutrophils, whereas the β chemokine MIP-1α has also been shown to be a powerful granulocyte-activating chemokine. ^30,39

In the present study we found that stimulation of the alveolar epithelial cell line A549 with a mixture of cytokines induced a massive release of IL-8, whereas the tracheal instillation of LPS in vivo was followed by a large increase in both MIP-2 and MIP-1α in the BALF 24 hours after stimulation. These effects were associated with a massive alveolar infiltration by leukocytes, notably polymorphonuclear cells, as evidenced by the high myeloperoxidase activity. A major finding was that inosine significantly downregulated the expression of IL-8 in vitro, as well as that of MIP-2 in vivo, while it marginally (P = .06) limited the production of MIP-1α. These protective effects were observed in relatively severe proinflammatory conditions, because we used both a maximal cytokine stimulation in vitro and a very high dose (50 μg) of LPS in vivo.

The corollary to these effects was a striking reduction in the number of leukocytes in the BALF and a marked decrease in myeloperoxidase activity in mice treated with inosine. Further, morphologic evaluation clearly showed that inosine was associated with a large reduction in the amount of leukocytes infiltrating the lungs. Although the molecular mechanisms underlying the effects of inosine on chemokine expression cannot be inferred from our data, it is noteworthy that we previously found that inosine reduced MIP-1α production in LPS-stimulated murine macrophages in vitro ¹⁶ and attenuated the rise of both MIP-2 and MIP-1α in lung and liver tissues in murine septic shock. These findings indicate that a major effect of inosine may be to alter the development of chemotactic gradients, thereby acting as a powerful downregulator of leukocyte trafficking in inflammatory conditions.

Inosine Affects the Cytokine Balance During Lung Inflammation

ARDS is associated with the development of interconnected inflammatory cascades, with proinflammatory cytokines playing a central role in the initiation and propagation of the inflammatory response leading to lung injury. In this respect, TNF-α and IL-1β are considered pivotal mediators of lung inflammation in ARDS. ^6,28 The release of TNF-α and IL-1β by alveolar macrophages triggers target cells (notably epithelial and endothelial cells) to produce additional mediators, thereby amplifying the initial inflammatory response. High BALF levels of TNF-α, IL-1β, and IL-6 are recovered from ARDS patients, and these levels are significantly higher in nonsurvivors than in survivors. ^40,41

In line with these concepts, we found that mice challenged with LPS expressed very large amounts of proinflammatory cytokines in their BALF. In contrast, treatment with inosine was associated with a significant reduction in the levels of all measured cytokines, whereas it also increased the production of the antiinflammatory cytokine IL-4 compared with sham animals. IL-4 has been shown to block the LPS-induced production of IL-1, IL-6, IL-12, IL-8, and TNF by monocytes ^42,43 and to inhibit the cytokine-induced expression of adhesion molecules (ICAM-1 and ELAM-1) on endothelial cells. ⁴⁴ This observation indicates that inosine is able to redirect the cytokine balance toward a more antiinflammatory profile, in line with previous data from our laboratory, where we found that inosine increased the production of IL-10, another counterregulatory cytokine, in endotoxemic mice. ¹⁶ Interestingly, we did not find such an effect of inosine on IL-10 in the present study, a difference that may be related to differences in the timing of inosine administration (before vs. after LPS) and the route of administration of LPS (local vs. systemic).

Inosine Reduces Lung Nitric Oxide Production and Reduces Nitrosative Stress After Lipopolysaccharide

The role of nitric oxide as a mediator of lung injury in ARDS remains a controversial issue. Nitric oxide may dampen the inflammatory response by inhibiting the activation of the transcription factor NFκB, and by reducing leukocyte interaction with the endothelium. ^26,45 On the other hand, enhanced nitric oxide formation may lead both to direct and indirect cytotoxic effects via the formation of peroxynitrite, a major oxidant species formed from the rapid reaction of nitric oxide with the superoxide anion. ²⁶ In this respect, it is noteworthy that mice genetically deficient in type II NOS gene are more resistant to LPS-induced acute lung injury. ⁴⁶ Mice challenged with LPS showed both an enhanced formation of nitric oxide and an intense immunostaining for 3-NT, an indirect marker of peroxynitrite generation. In contrast, inosine markedly suppressed 3-NT staining, an effect that can be ascribed to several mechanisms. First, by reducing nitric oxide production and the infiltration of leukocytes, which are important generators of O₂⁻, inosine is expected to limit the formation of peroxynitrite. Second, the metabolism of inosine results in the formation of uric acid (see below), a direct scavenger of oxyradicals and peroxynitrite. ^47–49 Our current findings suggest that a reduction in peroxynitrite-mediated cytotoxicity was an additional mechanism underlying the protective effects of inosine in murine ARDS.

Inosine Reduces High-Permeability Edema and Improves Lung Morphology

Widespread destruction of the alveolar epithelium and flooding of the alveolar spaces with proteinaceous exudates containing large amounts of neutrophils represent the typical lesion in ARDS. ⁵⁰ Mice exposed to LPS presented all these features, notably a high protein content of the BALF, attesting to the development of high-permeability pulmonary edema. This alteration was markedly prevented by inosine, indicating that this purine exerted a salutary effect on the integrity of the alveolocapillary membrane, as further substantiated by a marked improvement of lung morphology. The reduction in leukocyte infiltration was certainly an important mechanism underlying these beneficial effects, because neutrophils are considered as the primary cellular effectors of alveolocapillary damage in ARDS. ^2,3,27

Potential Mechanisms of Action of Inosine

The cellular and molecular mechanisms underlying the effects of inosine are incompletely understood. Previous investigations have determined that inosine acts, at least in part, via binding to the A3 adenosine receptor. ^14–16 In this respect, it is notable that A3 binding sites are abundantly present in lung homogenates from rodents, ¹⁴ and also that transcript for the A3 adenosine receptor is high in human lung. ⁵¹ Although part of the lung A3 receptors is attributable to resident mast cells, expression and function of A3 receptors in non-mast cells are unknown. On the basis of our data showing protective effects of inosine on human alveolar epithelial cells, we can speculate that these cells also express A3 adenosine receptors.

Recent data have indicated that inosine exerts cytoprotective effects in rat mast cells by stimulating the phosphorylation of protein kinase B (Akt). ¹⁷ In addition, inosine may partially act by interfering with the activation of the nuclear enzyme poly (ADP-ribose) polymerase (PARP). ⁵² Activation of PARP by oxidant-mediated DNA damage has been shown to be a major mechanism of tissue injury in various conditions, including inflammation, ischemia–reperfusion, and circulatory shock. ⁵³ Purines such as hypoxanthine, inosine, and adenosine share structural similarities with NAD, the substrate of PARP, and we have shown that these purines inhibit PARP activation in vitro. ⁵² Ultimately, inosine may also partially act via its breakdown product, uric acid, which has been consistently shown to be a scavenger of oxyradicals and of peroxynitrite. ^47–49 It appears, then, likely that inosine acts through multiple unrelated pathways, and future studies will be necessary to delineate the nature of these various pathways.

Clinical Relevance

The data presented in this study were obtained in a clinically relevant model of ARDS. Indeed, ARDS most frequently complicates the course of gram-negative sepsis, and LPS is currently considered to be a major eliciting factor in the development of ARDS in this setting. In addition, our observations were made while inosine was administered as a posttreatment strategy, which more precisely reflects the clinical scenario than any pretreatment intervention. The relatively high dose of inosine (3 × 200 mg/kg) used in this study was determined after pilot experiments where we found that at 3 × 100 mg/kg, inosine only partially reduced lung inflammation, and is consistent with the doses of inosine required to exert protective effects in various models of systemic inflammation. ^16,18 Inosine has a high safety profile (LD₅₀ > 20 g/kg), and we did not observe any particular side effects in the animals treated with this compound. Inosine has been safely given to humans for prolonged periods of time (up to 10 g/kg per day). ^54–56 It is possible that a local route of administration (inhalation therapy) might allow the use of much lower doses of inosine to achieve comparable antiinflammatory effects.

CONCLUSIONS

The data presented in this study indicate that inosine produces marked antiinflammatory effects in a clinically relevant model of ARDS. The current data, when coupled with recently emerging data showing reduced multiorgan dysfunction and an improved survival rate in the presence of inosine treatment in rodents subjected to endotoxin shock and polymicrobial sepsis, ^16,18,57 strengthen the possibility that inosine may have therapeutic utility in various surgical indications. Nevertheless, further investigations are required to determine the potential clinical usefulness of inosine in the adjunctive therapy of ARDS.