Azithromycin ameliorates sulfur dioxide-induced airway epithelial damage and inflammatory responses

Abstract

Background: The airway epithelium (AE) forms the first line of defence against harmful particles and pathogens. Barrier failure of the airway epithelium contributes to exacerbations of a range of lung diseases that are commonly treated with Azithromycin (AZM). In addition to its anti-bacterial function, AZM has immunomodulatory effects which are proposed to contribute to its clinical effectiveness. In vitro studies have shown the AE barrier-enhancing effects of AZM. The aim of this study was to analyze whether AE damage caused by inhalation of sulfur dioxide (SO₂) in a murine model could be reduced by pre-treatment with AZM.

Methods: The leakiness of the AE barrier was evaluated after SO₂ exposure by measuring levels of human serum albumin (HSA) in bronchoalveolar lavage fluid (BALF). Protein composition in BALF was also assessed and lung tissues were evaluated across treatments using histology and gene expression analysis.

Results: AZM pre-treatment (2 mg/kg p.o. 5 times/week for 2 weeks) resulted in reduced glutathione-S-transferases in BALF of SO₂ injured mice compared to control (without AZM treatment). AZM treated mice had increased intracellular vacuolization including lamellar bodies and a reduction in epithelial shedding after injury in addition to a dampened SO₂-induced inflammatory response.

Conclusions: Using a mouse model of AE barrier dysfunction we provide evidence for the protective effects of AZM in vivo, possibly through stabilizing the intracellular microenvironment and reducing inflammatory responses. Our data provide insight into the mechanisms contributing to the efficacy of AZM in the treatment of airway diseases.

Keywords: Azithromycin; Glutathione-S-transferase; Immunomodulation; Lamellar bodies; Lung barrier enhancement.

Fig. 1

Subtle damage to airway epithelium and increased leakiness occurs after SO₂ exposure which is mitigated by pretreatment with AZM. a. Schematic depiction of the SO₂ exposure system. 500 ppm SO₂ is downmixed to the desired concentration in a positive pressure control unit. Downmixed gas is passed over the animals in a tower, ensuring constant and comparable exposure to all animals. Up to 12 animals were treated simultaneously. Outflow gas was passed through a NaOH solution for scrubbing SO₂ gas from the air mixture. Outflow pressure was monitored in order to maintain a small positive pressure (1 cm H₂O) in the exposure tower. b. Weight changes in mice exposed to 100 ppm SO₂ for 4 h. Overall, mice placed in the tower lost a significant (8%; *** p < 0.0001) amount of body weight after exposure to SO₂ in the inhalation system. This weight loss did not occur in control animals. Weight loss was recovered quickly, and mice reached their starting weights within a week after exposure. Data are means of ± SD of n = 18 for SO₂ exposed mice and n = 4 for control mice. c. No histological changes were observed after exposure to SO₂ H&E stained lung tissue. No histological changes were observed. Scale bars are 500 μm. d. HSA in BALF is increased upon SO₂ exposure. Increased HSA in SO₂ exposed mice compared with control mice was seen, as determined by ELISA. Average HSA in control and SO₂ treated mice was 143 and 440 ng/ml, respectively. Data are represented as box and whiskers plots with boxes encompassing the 25th – 75th quartiles, and whiskers demonstrating minimum and maximum values. n = 5 mice in each group. p = 0.0044. e. AZM pre-treatment results in decreased epithelial permeability to HSA. Mice orally gavaged with 2 mg/kg AZM 5 times per week had a trend for reduced HSA concentration in BALF observed after SO₂ exposure when compared to placebo mice, indicating a protective effect. Average HSA in control, SO₂ and AZM, SO₂ groups was 238, 350 and 150 ng/ml, respectively. Data are represented as box and whiskers plots with boxes encompassing the 25th – 75th quartiles, and whiskers minimum and maximum values. n = 5 observations per treatment group. p = 0.13

Fig. 2

SO₂ exposure results in increased expression of members of the Glutathione S Transferase family. a. Glutathione S transferase is increased in BALF in SO₂ exposed mice. BALF samples were analysed by mass spectrometry. Of the samples showing dramatic changes from controls, 6 GST analogs were detected and demonstrated to be increased in BALF samples of SO₂ exposed mice. Data represent a single measurement from a pooled sample from six mice in each treatment group. b. The increased GST concentration in BALF seen in SO₂ exposed mice is attenuated by pre-treatment with AZM. The six GSTs shown previously all had reduced BALF concentrations when mice were pre-treated with AZM. Data represent a single measurement from a pooled sample from six mice in each treatment group

Fig. 3

Pretreatment with AZM reduces the expression of SO₂ induced inflammatory interferon alpha responsive genes. a. Gene set enrichment analysis of positively correlated pathways with the highest normalized enrichment scores (NES) as a result of SO₂ challenging of the mice. Top three most positively correlated pathways were the inflammation related interferon alpha response, interferon gamma response and allograft rejection. FDR is false discovery rate and NOM p-val is nominal p-value. b. Volcano plot of differentially expressed genes comparing gene expression of non-exposed vs SO₂ exposed mice. Genes in the interferon alpha response pathway are colored green. c. Volcano plot of differentially expressed genes comparing gene expression of non-exposed vs SO₂ exposed mice that had been pre-treated with AZM. Genes in the interferon alpha response pathway are colored green. d. Heatmap of genes that were positively correlated to the interferon alpha gene set. On the left is shown the differential gene expression as a result of SO₂ exposure. On the right is shown the differential expression as result of SO₂ exposure in AZM pre-treated mice. e. Boxplot of selected genes in the interferon alpha gene set. All genes show a similar pattern of higher expression as a result of SO₂ exposure. The AZM pre-treatment of the unexposed as well as the SO₂ exposed resulted in similar expression levels as the unexposed, untreated mice. Boxplot made with the R package ggplot/geom. Data are represented as box and whiskers plots with boxes encompassing the 25th – 75th quartiles, and whiskers demonstrating 1.5 Interquartile range above and below. N = 3, except for placebo (N = 2). *** p ≤ 0.001

Fig. 4

AZM pre-treatment protects against epithelial shedding and increases lamellar body formation. a. Electron microscope images of the tracheal epithelium. SO₂ exposed mice show shedding of the epithelium as seen when comparing SO₂ +/−. Shedding is seen by the top of the layer peeling of and an uneven layer indicated by red arrows. Mice that had been pre-treated with AZM, but not exposed to SO₂ showed accumulations of vesicles in the epithelial layer. SO₂ exposed, AZM pre-treated mice also showed these vesicle formations, even to a greater extent, and the epithelial layer did not show shedding. All the epithelial layers presented some small vesicle formations, but the increases as a result of AZM treatment were distinct throughout all the samples. Yellow arrows point to examples of vesicle formations. Shown are representative images taken of 3 replicates. Top scale bars are 10 μm and lower scale bars are 2 μm. b. Electron microscope images of the alveolar region of the lungs. No discernible difference was seen in the control vs the SO₂ exposed mice. Lamellar bodies were observed in most of the alveolar type II cells in all the samples. Only in the AZM pre-treated mouse samples were accumulations of lamellar bodies detected. Yellow arrows point to examples of lamellar bodies. Shown are representative images taken of 3 replicates. Top scale bars are 10 μm and lower scale bars are 2 μm