Azithromycin Improves Macrophage Phagocytic ... - ATS Journals

Azithromycin Improves Macrophage Phagocytic Function and Expression of Mannose Receptor in Chronic Obstructive Pulmonary Disease Sandra Hodge1,2, Greg Hodge1,2, Hubertus Jersmann1,2, Geoffrey Matthews1, Jessica Ahern1, Mark Holmes1,2, and Paul N. Reynolds1,2 1

Department of Thoracic Medicine, Royal Adelaide Hospital and Lung Research Laboratory, Hanson Institute, Adelaide, South Australia, Australia; and 2University of Adelaide, Adelaide, South Australia, Australia

Rationale: Defective efferocytosis (phagocytic clearance of apoptotic cells) in the airway may perpetuate inflammation via secondary necrosis in chronic obstructive pulmonary disease (COPD). We have previously reported that low-dose azithromycin improved alveolar macrophage (AM) phagocytic function in vitro. Objectives: We investigated collectins (mannose-binding lectin [MBL] and surfactant protein [SP]-D) and mannose receptor (MR) in COPD and their possible role in the azithromycin-mediated improvement in phagocytosis. Methods: In vitro effects of azithromycin on AM expression of MR were investigated. MBL, SP-D, and MR were measured in patients with COPD and control subjects. Azithromycin (250 mg orally daily for 5 d then twice weekly for 12 wk) was administered to 11 patients with COPD. Assessments included AM phagocytic ability and expression of MR, MBL, SP-D, bronchial epithelial cell apoptosis, pulmonary function, C-reactive protein, blood/BAL leukocyte counts, cytokine production, and T-cell markers of activation and phenotype. Measurements and Main Results: Azithomycin (500 ng/ml) increased MR expression by 50% in vitro. AM MR expression and levels of MBL and SP-D were significantly reduced in patients with COPD compared with control subjects. In patients with COPD, after azithromycin therapy, we observed significantly improved AM phagocytic ability (pre: 9.9%; post: 15.1%), reduced bronchial epithelial cell apoptosis (pre: 30.0%; post: 19.7%), and increased MR and reduced inflammatory markers in the peripheral blood. These findings implicate the MR in the defective phagocytic function of AMs in COPD and as a target for the azithromycin-mediated improvement in phagocytic ability. Conclusions: Our findings indicate a novel approach to supplement existing therapies in COPD. Keywords: chronic obstructive pulmonary disease; alveolar macrophage; phagocytosis; azithromycin; apoptosis

Chronic obstructive pulmonary disease (COPD) is a high-burden disease and is predicted to become the third most common cause of death in the world by 2020 (1). Extensive public health campaigns and pharmaceutical approaches to promote smoking cessation have been implemented; however, a large percentage of the adult population continues to smoke and risks developing COPD. Furthermore, many patients only stop smoking once

(Received in original form November 11, 2007; accepted in final form April 16, 2008) Supported by a National Health and Medical Research Council Project Grant, by an NHMRC Clinical Career Development Award and Practitioner Fellowship, and by a Pfizer and Royal Adelaide Hospital Clinical Project Grant. Correspondence and requests for reprints should be addressed to Dr. Sandra Hodge, M.Sc. Ph.D., Lung Research, Hanson Institute, Frome Road, Adelaide, South Australia 5001. E-mail: [email protected] This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org Am J Respir Crit Care Med Vol 178. pp 139–148, 2008 Originally Published in Press as DOI: 10.1164/rccm.200711-1666OC on April 17, 2008 Internet address: www.atsjournals.org

AT A GLANCE COMMENTARY Scientific Knowledge on the Subject

Defective phagocytosis in the airway may perpetuate inflammation in chronic obstructive pulmonary disease (COPD). Low-dose azithromycin improves phagocytosis in vitro; however, the mechanisms and the effect of in vivo administration to subjects with COPD have not been determined. What This Study Adds to the Field

Our findings of improved macrophage phagocytic ability and reduced systemic inflammation after low-dose azithromycin therapy provide further rationale for investigating macrolides as supplements to existing therapies in COPD.

COPD is established and therefore need effective ongoing treatment. Current treatment strategies for COPD generally have limited efficacy. Therefore, we believe that further improvement requires a new treatment focus. We have conducted a number of studies investigating the role of apoptosis and macrophage dysfunction in the airways in COPD. We have described the pathological concept of dysregulated apoptosis of bronchial epithelial cells and defective phagocytosis of apoptotic cells by alveolar macrophages (AMs) in the airways of subjects with COPD (i.e., defective efferocytosis, a term used to differentiate phagocytosis of apoptotic cells from the more widely studied complement or Fcg receptor– mediated phagocytosis) (2–5). We have shown a net increase in apoptotic material and evidence of secondary necrosis (increased lactate dehydrogenase) in the airway in COPD that have the potential to perpetuate an inflammatory response (3). Our current research is focusing on identifying and implementing treatment strategies specifically aimed at improving efferocytosis in COPD. A number of studies, including those in panbronchiolitis and cystic fibrosis (CF), have shown that low-dose macrolide antibiotics, including azithromycin, have antiinflammatory properties and clinical benefits (6, 7). Such antiinflammatory effects may be enhanced by the ability of azithromycin to reach very high concentrations in AMs (to a greater extent than other macrolide antibiotics) with sustained concentrations in tissues (8). In addition to the antiinflammatory properties, we have reported that azithromycin, when used in subbactericidal concentrations, improves the phagocytic function of AMs taken from the lungs of subjects with COPD ex vivo, although the mechanisms for this effect are unknown (9). The lung collectins are a related group of collagenous lectins (primarily surfactant protein [SP]-A, SP-D, and mannose bind-

140

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 178

2008

ing lectin [MBL]), which, along with the collectin-like molecule C1q, have important roles in mediating host defense via the facilitation of macrophage phagocytosis of invading pathogens (10, 11). All of these molecules have a collagenous tail at one end and a globular carbohydrate recognition domain at the other. MBL is primarily produced in the liver and is not seen in ‘‘resting’’ lung but is up-regulated in pulmonary inflammation. SP-A and SP-D are produced by alveolar type II cells and by Clara cells. These molecules also play a key role in regulating the clearance of apoptotic cells via several receptors that bind to their collagenous tail (including the lung collectin associated receptor, LDL receptor-related protein [CD91]–calreticulin complex) and in the regulation of pulmonary inflammatory responses (11, 12). SP-D deficiency has been associated with the development of emphysematous changes and a reduction in apoptotic cell clearance in murine models (12). The present study aimed to identify the mechanisms for the azithromycinmediated improvement in AM phagocytic ability, primarily by focusing on the collectin pathway. Having determined a link between the effects of azithromycin and the collectin pathway, we investigated whether the use of low-dose azithromycin has clinical benefits in the treatment of COPD. Specifically, we aimed to evaluate the effect of low-dose oral azithromycin for 12 weeks on efferocytosis in COPD using cells derived from bronchoalveolar lavage (BAL). The study was open-label, uncontrolled, and primarily focused on objective biological responses obtained from the bronchoscopy samples taken. Primary outcomes were AM phagocytic ability and epithelial cell apoptosis in the airways of subjects with COPD. The effects of azithromycin on collectins, mannose receptor (MR), lung function, quality of life, and inflammatory mediators were assessed. Some of the results of these studies have been communicated in part at a recent international meeting and have been presented in abstract form (13).

4weeks before the first screening visit and throughout the study; treatment with oxygen therapy or nocturnal positive pressure for sleep apnea, treatment with more than 10 g/day of prednisolone (or equivalent); allergy to macrolide antibiotics; current treatment with medications known to significantly interact with azithromycin (including antacids and ergot preparations); prolonged QT interval (.0.42 s); and drugs associated with prolongation of QT interval, pregnancy or lactation, or significant renal impairment (estimated glomerular filtration rate, ,30 ml/m). Withdrawal criteria included request by the patient, any serious adverse event and any major exacerbation requiring the use of azithromycin at standard doses, development of any serious concurrent illness, or cessation of cigarette smoking among the group of currentsmoker subjects with COPD. For in vitro studies and measurement of MBL, SP-D, and MR, 10 healthy adult volunteers were recruited. These subjects were never-smokers with normal lung function (four women, six men; FEV1 [%pred] [mean 6 SEM], 101 6 3; FEV1/FVC, 85 6 3; age, 43 6 5 yr; BAL yield, 75 6 5 ml).

METHODS

Collection of Breath Condensate

Azithromycin (Zithromax) was purchased from Pfizer, Inc. (New York, NY). The following monoclonal antibodies and immunological reagents were used. For identification of leukocyte subsets: CD45 phycoerythrin cyanide-5 (PC-5), CD14 PC-5, CD3 PC-5, and CD33 PC-5. For measurement of AM recognition molecules. MR phycoerythrin (PE), and CD36 (fluorescein isothiocyanate [FITC]) (Immunotech/Coulter, Marseille, France); CD44 (FITC) (BD Biosciences, San Jose, CA); CD31 and CD91 (Serotec, Oxford, UK), for enumeration of regulatory T cells (Tregs); Th1, cytotoxic and activated T cells: CD8 (FITC), CD25 (FITC), CD127 (PE), CD28 (PE), 62L (FITC), CD45RO (PE) (BD Biosciences); for investigation of apoptosis: 7amino-actinomycin D (7-AAD) (Sigma Chemicals, Castle Hill, Australia); for phagocytosis: Mitotracker red (MTR; Molecular Probes, OR). Cytometric bead array (CBA) kits were supplied by BD Biosciences. For receptor blocking, anti-human mannose receptor was purchased from HyCult biotechnology (Uden, The Netherlands). MBL was obtained from R&D (Minneapolis, MN).

Subject Population Eleven patients with COPD (five current smokers and six ex-smokers > 1 yr) were included. Approval was granted by the Royal Adelaide Hospital Ethics Committee, and written informed consent was obtained for each patient or control subject recruited for the study. The diagnosis of COPD was established using the GOLD (Global Initiative for Chronic Obstructive Lung Disease) criteria (FEV1/FVC , 70%) with clinical correlation. Subjects were between 40 and 75 years of age. Patient demographics are presented in Table 1. Exclusion criteria included the following: past or present disease (which, as judged by the investigator, could affect the outcome of the study); FEV1 of less than 1.4 L (for ethical reasons, research bronchoscopy is not performed on subjects with an FEV1 , 1.4 L); diagnosis of other lung disease, cardiovascular disease, or malignancy; respiratory tract infection in the

Flexible Bronchoscopy Flexible bronchoscopy was performed according to expert consensus recommendations for the performance of bronchoscopy for investigative purposes (American Thoracic Society) and as previously reported by us (2, 3, 5, 9). Further details are provided in the online supplement.

Blood Collection Venous blood was collected into tubes containing 10 U/ml preservative-free sodium heparin (DBL, Sydney, Australia). Blood differential cell counts were performed using a Cell Dyn 4000 (Abbott Diagnostics, Sydney, Australia). Total cell counts in BAL were performed using a modified Neubauer hemocytometer. High-sensitivity C-reactive protein (CRP), blood, and BAL differential cell counts and blood biochemistry were measured by the Division of Clinical Pathology (Institute of Medical and Veterinary Science, Adelaide, Australia) using an Olympus AU-5400 analyzer (Olympus Optical Co. Ltd, Shunjoku, Japan). Samples were processed for flow cytometry within 1 hour of collection. For subsequent cytokine analysis, samples were centrifuged at 3,000 3 g for 5 minutes, and plasma was removed and frozen at 2708C.

Breath condensate was collected at using a commercial exhaled breath condenser collector (RTube; Respiratory Research, Inc., Charlottesville, VA). Samples were immediately stored at 2708C.

Preparation of Samples BAL and bronchial brushing–derived cells were washed in Gibco RPMI 1640 (Invitrogen, Mount Waverley, Australia), and re-suspended in culture medium supplemented with 10% fetal calf serum (Gibco; Invitrogen) at a concentration of 4 3 105/ml. TABLE 1. CLINICAL CHARACTERISTICS OF SUBJECTS WITH CHRONIC OBSTRUCTIVE PULMONARY DISEASE Patient Pack- INH. St. FEV1 FEV1/ BAL Volume (ml), Number Age Smoker years (yes/no) Sex (%pred) FVC Bronch 1, 2 1 2 3 4 5 6 7 8 9 10 11 Mean SEM

62 57 65 42 57 68 71 61 72 60 65 62 2.5

Ex Curr Curr Curr Curr Ex Ex Ex Ex Ex Curr

62 50 75 20 100 40 30 170 35 60 50 62 112.6

Y N N N Y Y N N N N N

M F M F M M M M M M M

58 101 47 76 34 59 74 91 47 89 50 66 6.5

37 65 50 66 27 52 66 64 36 59 44 51 4.1

30, 72, 59, 75, 15, 30, 65, 75, 16, 80, 38,

25 65 42 20 15 30 59 42 20 84 63

Definition of abbreviations: BAL volume Bronch 1, 2 5 bronchoalveolar lavage return (ml) at first and second bronchoscopy; COPD 5 chronic obstructive pulmonary disease; Curr 5 current smoker; Ex 5 ex-smoker; INH. St. 5 inhaled steroids.

Hodge, Hodge, Jersmann, et al.: Azithromycin Increases Phagocytosis In Vivo

Phagocytosis of Apoptotic Bronchial Epithelial Cells by AMs Immortalized normal bronchial epithelial cells (16HBE) were used as phagocytic targets for the phagocytosis assay. The cell line was maintained and passaged as previously described (2). For use in the phagocytosis assay, the cells were induced to apoptosis using ultraviolet radiation for 20 minutes and stained with MTR as previously described (2, 9). Flow cytometry was applied to measure phagocytosis of the MTR-stained apoptotic bronchial epithelial cells by AMs after 90 minutes as previously reported (2, 9).

In Vitro Studies This in vitro phase of the study was designed to investigate the possible functional link between azithromycin activities, the collectin pathway, and AM phagocytic ability. AMs were prepared from BAL collected from five never-smoker control subjects as described previously. Phagocytosis of apoptotic cells after blocking of MR or addition of MBL to AMs in vitro. To investigate the effects of MBL or reduced expression of MR on AM phagocytic ability, AMs collected from five never-smoker control subjects were incubated with anti-MR (10 mg/ml) or control IgG1 for 30 minutes at 378C before the addition of apoptotic cells, and the phagocytosis assay was performed as described previously. Separate experiments were performed in the presence of 2, 10, and 20 mg/ml MBL. Effects of azithromycin on expression of MR by AMs in vitro. We investigated the effects of azithromycin on expression of MR in AMs collected from 5 of the 10 never-smoker control subjects. AMs were exposed to azithromycin at concentrations equivalent to low dose (500 ng/ml) to high dose (10,000 ng/ml) for 48 hours as previously reported (9). Expression of intracellular MR was assessed using flow cytometry as previously published (5). The permeabilization procedure allows for investigation of surface and intracellular MR. This was performed because of reports of trafficking of preformed MR to the cell surface and because in differentiated macrophages, 80% of the MR is localized intracellularly in vesicles and the intracellular pool may include newly synthesized receptors en route to the endosomal apparatus and receptors moving from one pool to another (14, 15). A further report showed that intracellular MRs that are recycled to the surface are involved in phagocytosis of lipoarabinomannan-coated microspheres (16).

Collectins and MR in BAL from Patients with COPD and Control Subjects ELISA measurement of MBL and SP-D in healthy control subjects and patients with COPD. MBL and SP-D were measured in BAL that had been stored at 2708C. To increase the power of this study and to investigate the effect of smoking among patients with COPD, we included stored samples from a further four current and six ex-smoking patients with COPD from a previous study (5) (age [mean 6 SEM], 63 6 2 yr; FEV1 [% predicted], 62 6 3; FEV1/FVC [%], 58 6 4; smoking, 58 6 5 pack-years) in addition to subjects with COPD from the present study. The same subjects were investigated for MBL and SP-D. The individuals who were excluded from the azithromycin treatment study were not included. ELISAs (MBL: Hycult Biotechnology, Uden, The Netherlands; and SP-D: Biovendor, Modrice, Czech Republic) were performed following the instructions supplied by the manufacturer. Flow cytometric analysis of MR in healthy control subjects and patients with COPD. We applied flow cytometry to analyze the percentage of AMs expressing MR in BAL from the nine patients with COPD (out of 11) who were included in the treatment study and the one patient with COPD who ceased smoking after his first bronchoscopy (six current smokers and four ex-smokers) and compared the expression to the group of 10 healthy never-smoker control subjects.

Administration of Azithromycin to Patients with COPD To further investigate the azithromycin-mediated improvement in phagocytic ability, we administered azithromycin at low dose to a cohort of subjects with COPD. The study was open-label and uncontrolled and primarily focused on objective biological responses obtained from the bronchoscopy samples. Screening visit. Eleven patients with COPD were included as described previously and in Table 1. A screening visit was performed

141

within 14 days of the first dosing occasion. At this visit, subjects underwent physical examination, lung function and electrocardiogram (ECG) testing, chest X-ray and laboratory tests (liver enzymes, electrolytes and serum creatinine, complete blood picture), and highly sensitive CRP testing. On this occasion, a determination was made by a thoracic medicine physician as to whether the patient’s baseline therapy conformed to current best practice. If not, it was adjusted accordingly. Subjects were required to be on a stable medication regimen for 3 months before entering the study. Treatment phase. After screening, a flexible bronchoscopy (FB) was performed, and baseline samples were collected. Treatment was commenced with azithromycin 250 mg orally daily for 5 days then twice weekly for a total of 12 weeks. Subjects underwent repeat physical examinations, pulmonary function and ECG testing, and blood and breath condensate assessments every 4 weeks throughout the study. A repeat FB was performed during Weeks 12 to 13 (while the patient was taking azithromycin). After the 12-week treatment phase, azithromycin dosing ceased. The subjects were assessed every 4 weeks for an additional 12 weeks. Study assessments. Key assessments were AM function, bronchial epithelial cell apoptosis, and collectin-associated mediators. Other assessments included lung function, quality-of-life monitoring, and BAL and inflammatory markers in BAL blood and breath condensate samples as outlined below. The influence of current smoking and disease severity was assessed. BAL was processed for bacterial culture to determine whether bacterial colonization was associated with response to azithromycin. Phagocytic ability of AMs. AMs were collected from BAL and assessed for their ability to phagocytose apoptotic bronchial epithelial cells as described previously and as reported (2, 5, 9). Apoptosis of bronchial epithelial cells. Apoptosis of bronchial epithelial cells (from bronchial brushings collected before and after azithromycin therapy) was assessed using flow cytometry and staining with 7-AAD as previously reported (3). MBL, SP-D, MR, and other AM recognition molecules. MBL and SP-D were assessed on BAL samples before and after therapy using ELISA as described previously. We further investigated the effects of azithromycin therapy on MR. We have previously shown significantly reduced AM expression of platelet endothelial cell adhesion molecule (CD31) and LDL receptor–related protein (CD91) in COPD (5). These receptors are important in the recognition of apoptotic cells by AMs; therefore, we quantified expression of these markers and of thrombospondin receptor (TSP-R, CD36) on AMs obtained from BAL before and after azithromycin therapy using flow cytometric methods as previously reported (5). In view of our findings and reports by others of enhancement of phagocytic function by engagement of the hyaluron receptor (CD44) on AMs (9), we also investigated the effects of lowdose azithromycin therapy on AM expression of CD44. Pulmonary function and ECG. Pre- and post-bronchodilator spirometry was performed at the screening assessment and at 4-week intervals. These assessments continued for 12 weeks after cessation of azithromycin treatment. Because a possible side effect of azithromycin is the prolongation of QT interval, ECGs were performed using a GE Marquette MAC 500 ECG System (Harrell Medical, Inc., Lake Oswego, OR). St George’s Respiratory Questionnaire. Quality-of-life assessment was performed using the St George’s Respiratory Questionnaire. For details, see the online supplement. Proinflammatory cytokines. Breath condensate and blood was collected at 4-week intervals from commencement of azithromycin therapy to 12 weeks after treatment. Condensate, BAL, and plasma were stored at 2708C and tested in batches. Inflammatory cytokines tumor necrosis factor (TNF)-a, IL-6, IL-1b, and IL-8 were quantified in plasma, breath condensate, and BAL before and after azithromycin treatment using an inflammatory CBA kit and CBA software (BD Biosciences) as previously reported (17). Leukocyte subtypes, activated T cells, and Th1, Tregs, and cytotoxic T cells. For peripheral blood and BAL, the total leukocyte count (WCC) and the percentage and absolute numbers of lymphocytes, monocytes, and neutrophils were recorded. The percentage of CD31 and CD41/ CD81 T cells were calculated in blood, BAL, and bronchial brushings (intraepithelial T cells) using flow cytometry as previously described (18). Samples were analyzed using a panel of fluorescently conjugated

142


monoclonal antibodies to further characterize leukocyte subtypes and a combination of conjugated and unconjugated monoclonal antibodies to assess surface expression of receptors. Enumeration of Tregs (CD31CD251CD1272) (19), Th1 (CD31CD62L2CD45RO1) (20), and cytotoxic (CD81CD281) and activated T cells (CD31CD251) was performed on samples of blood, brush (intraepithelial T cells), and BAL. Markers of oxidative stress. GSH (reduced glutathione) availability was determined in BAL and breath condensate as we have previously published, with minor modifications (21). H2O2 concentration in BAL and breath condensate was assessed using the method of Jackson and colleagues (22). 8-Isoprostane levels in BAL and breath condensate were determined according to the manufacturer’s instructions (Cayman Chemicals, Ann Arbor, MI). For details, see the online supplement.

2008

specific for MR before assessing phagocytic ability. Phagocytosis of apoptotic cells by AMs was decreased by 60 6 25% after blocking MR. MBL added to AMs before phagocytosis assay resulted in a dose-dependent increase in AM phagocytic ability (Figure 1B). Collectins and MR in BAL from Healthy Control Subjects and Current and Ex-smoker Patients with COPD

ELISA measurement of MBL and SP-D. Levels of MBL and SPD were significantly reduced in current- and ex-smoker subjects with COPD compared with healthy control subjects (Figures 2A and 2B). Levels of MBL were increased in the BAL of subjects with COPD who had ceased smoking compared with those who were currently smoking, although the difference did not reach statistical significance (P 5 0.06).

Data Analysis and Statistical Considerations The primary objective of this study was to ascertain whether there is evidence of an impact of low-dose azithromycin on macrophage function as a prelude to a subsequent placebo-controlled study. The null hypothesis was that there would be no significant effects of azithromycin treatment for any of the parameters tested. The alternative hypothesis was that there would be a significant difference in at least one tested parameter after treatment with azithromycin. The Wilcoxon signed rank test was applied to analyze the paired data. Significance was established as P , 0.05. The study was adequately powered to detect a difference of 20% in phagocytosis of apoptotic airway epithelial cells by macrophages and between subjects with COPD before and after azithromycin treatment. Variability estimates were taken from our previous studies of apoptosis, phagocytosis, and AM surface marker expression in patients with COPD (2, 3, 5, 9). Correlations between the phagocytic ability of AMs and apoptosis of bronchial epithelial cells, quality-of-life score, CRP, and blood WCC were performed using the Spearman’s rank test. This test was also applied to investigate correlations between age and SP-D, MBL, and MR. Statistical analyses were performed using SPPS, version 11 (SPSS, Inc., Chicago, IL).

RESULTS In Vitro Studies

Effects of azithromycin on expression of MR by AMs in vitro. We investigated the effects of azithromycin on AM expression of MR. The percentage of AMs expressing MR was significantly increased after exposure to low-dose (500 ng/ml) or high-dose (10,000 ng/ml) azithromycin for 48 hours (Figure 1A). Phagocytosis of apoptotic cells after blocking of MR or addition of MBL to AMs in vitro. AMs were treated with a blocking antibody

Flow Cytometric Analysis of MR

The percentage of AMs expressing MR was significantly lower (by 42%) in subjects with COPD compared with healthy neversmoker control subjects (Figure 2C). Administration of Azithromycin to Subjects with COPD

Subject population. Of the 11 COPD subjects included in the study, two were excluded from final analysis, after completing the study. One subject (number 11) ceased cigarette smoking immediately after his first bronchoscopy; therefore, any improvements in macrophage function could conceivably have been attributed to this variable. Another patient (number 4) was excluded based on persistent growth of Haemophilus influenzae and Pseudomonas aeruginosa in BAL. Thus, nine subjects that completed the azithromycin course were included in the pre- versus post-analyses. Effect of azithromycin therapy on phagocytosis of apoptotic bronchial epithelial cells by AMs. There was a significant improvement in the ability of AMs to phagocytose apoptotic bronchial epithelial cells after 12 weeks of therapy with lowdose azithromycin (P 5 0.01) (Figure 3). Effect of azithromycin therapy on apoptosis of bronchial epithelial cells. There was a significant decrease in the percentage of apoptotic bronchial epithelial cells collected by bronchial brushing at FOB after 12-week therapy with low-dose azithromycin (P 5 0.01) (Figure 4). Effect of azithromycin therapy on MBL, SP-D, MR, and other AM recognition molecules. There were no significant changes in

Figure 1. In vitro studies. (A) Effects of the addition of azithromycin (low dose, 500 ng/ml) or high-dose (10,000 ng/ ml) on the percentage of alveolar macrophages (AMs) expressing mannose receptor (MR). (B) Effects of (i) blocking MR on AMs and (ii) addition of mannose-binding lectin (MBL) on phagocytosis of apoptotic bronchial epithelial cells by AMs. *Significantly (P , 0.05) reduced % expression of MR in the presence of azithromycin, significantly reduced AM phagocytic ability after blocking of MR, and dose-dependent significant increase in AM phagocytic ability with MBL. Figures present data (mean 6 SEM) from three independent experiments.


Figure 2. Expression of collectins and mannose receptor (MR) in control subjects and current- and ex-smoker patients with chronic obstructive pulmonary disease (COPD). (A) Mannose-binding lectin (MBL) in bronchoalveolar lavage (BAL). (B) Surfactant protein (SP)-D in BAL. (C) Percentage of alveolar macrophages (AMs) expressing MR. Box plots present median 6 25th and 75th percentiles (solid box) with the 10th and 90th percentiles shown by whiskers outside the box. MBL and SP-D were investigated by ELISA in stored samples of BAL from 10 in each group of current- and ex-smoker subjects with COPD and neversmoker control subjects. MR was measured by flow cytometry in AMs collected by BAL from the 10 never-smoker control subjects, the nine recruited subjects with COPD who were included in the treatment study, and the one patient with COPD who ceased smoking after his first bronchoscopy (six current-smokers and four ex-smokers). *Significantly (P , 0.05) lower expression of MBL, SP-D, and MR in both COPD groups compared with control subjects.

BAL levels of MBL or SP-D after azithromycin treatment (data not shown). There was a significant increase in the expression of MR in AMs after azithromycin treatment (Figure 5). There was a nonsignificant trend for increased expression of AM recognition molecules CD31 and CD91 (P 5 0.123 and 0.161, respectively; data not shown) after 12 weeks of treatment with low-dose azithromycin. There were no significant changes in AM expression of CD36 or CD44 after azithromycin therapy (data not shown). Effect of azithromycin therapy on pulmonary function and ECG testing. There were no significant changes in spirometry

143

(pre- and post-bronchodilator) measured at the screening assessment and at 4-week intervals for any patients enrolled in the study (data not shown). There were no significant changes in QT interval measured at the screening assessment and at 4-week intervals thereafter (data not shown). Effect of azithromycin therapy on patient-assessed quality of life. The study was not powered to detect improvement in quality of life, but no overt deterioration was detected using the St. George’s Respiratory Questionnaire. Total median scores were as follows: pre, 48.8 (range, 1.6–69.7) versus post, 26.8 (range, 0.7–70.3) (P 5 0.26). Calculation of the three separate components (symptoms, activity, and impact) also revealed a trend for improvement after azithromycin therapy: symptom score pre, 31.7 (range, 9.8–86.3) versus post, 33.1 (range, 4.4– 73.4) (P 5 0.13); activity score pre, 41.8 (range, 0–93.9) versus post, 41.8 (range, 0–85.8) (P 5 0.31); and impact score pre, 24.8 (range, 0–62.9) versus post 15.1 (range, 0–60.5) (P 5 0.16). Effect of azithromycin therapy on inflammatory markers. DIFFERENTIAL CELL COUNTS AND BLOOD BIOCHEMISTRY. There was a significant decrease in the total WCC in peripheral blood after 12 weeks of therapy with low-dose azithromycin (P 5 0.01) (Figure 6A). This was associated with a nonsignificant decrease in lymphocyte and neutrophil numbers (data not shown). There were no significant changes in any biochemical parameters (e.g., liver enzymes, serum creatinine), and there were no significant changes in BAL volume, WCC, differential leukocyte counts, or the percentage of T cells or CD81 T cells in BAL collected before or after azithromycin treatment. In bronchial brushings, there were no significant differences in CD31 intraepithelial T cells or the percentage of CD81 T cells in samples collected before or after azithromycin (data not shown). CRP. There was a decrease in CRP in blood samples collected directly after azithromycin treatment versus baseline. Significantly decreased levels persisted for 4 weeks after cessation of therapy (P 5 0.04) (Figure 6B). INFLAMMATORY CYTOKINES. CBA analysis revealed a nonsignificant trend for decreased levels of IL-6, TNF-a, IL-b, and IL8 in blood, BAL, and breath condensate after azithromycin treatment (Table 2). There was a significant decrease in the level of IL-1b in BAL after azithromycin therapy (P 5 0.046) (Table 2). Effect of azithromycin therapy on markers of Th1 and activated T cells, Tregs, and Th1. In blood, BAL, and bronchial brushings, there were no significant changes in the percentage of T cells expressing markers of activation (CD25), cytotoxic (CD81CD281), or Th1 type (62l2CD45RO1) or Tregs (CD1272CD251) (Table 3). Effect of azithromycin therapy on markers of oxidative stress. There was a nonsignificant trend for reduced levels of H2O2 and 8-isoprostane and increased GSH in BAL or breath condensate samples after azithromycin treatment (median H2O2 in BAL: preazithromycin, 1.43 mM [range, 0.36–12.53] vs. postazithromycin, 1.20 mM [range, 0.32–2.99], not significant; H2O2 in breath condensate: preazithromycin, 1.70 mM [range, 0.57–5.23] vs. postazithromycin, 1.5 mM [range, 0.66–3.11], not significant; GSH in BAL: preazithromycin, 0.92 mM [range, 0.35–3.1] vs. postazithromycin, 0.74 mM [range, 0.33–2.2], not significant; GSH in breath condensate: preazithromycin, 0.07 mM [range, 0.01–0.1] vs. postazithromycin, 0.95 mM [range, 0.07–0.12], not significant; 8-isoprostane in BAL: preazithromycin, 17.0 pg/ml [range, 3.9–51.1] vs. postazithromycin 16.0 pg/ml [range, 1.1– 39.8], not significant. 8-Isoprostane was not detected in breath condensate from any patient. Correlations between inflammatory markers, AM phagocytosis, and apoptosis. There were significant negative correlations

144


2008

Figure 3. Effect of azithromycin treatment on phagocytosis of apoptotic bronchial epithelial cells by alveolar macrophages (AMs). Phagocytic ability of broncholaveolar lavage–derived AMs from subjects with chronic obstructive pulmonary disease (COPD) was assessed before and after treatment with 250 mg azithromycin twice weekly for 12 weeks. (A) Broken lines and solid lines represent current-smoker and ex-smoker subjects with COPD, respectively. (B) Data summarized in box plots as described in Figure 2. *Significant increase in AM phagocytic ability after azithromycin treatment compared with before treatment.

between blood WCC and the percentage of AMs that phagocytosed apoptotic cells and between blood WCC and bronchial epithelial cell apoptosis (r 5 20.567, P 5 0.022, and r 5 20.527, P 5 0.036, respectively). There were no significant correlations between AM phagocytosis and quality of life, CRP, or apoptosis of bronchial epithelial cells. There were no significant correlations between smoking status or disease severity and any of the parameters tested. Correlations between age and MBL, SP-D, and MR. There were no significant correlations between age and SP-D or MR. There was a weak but significant negative correlation between age and MBL (r 5 20.365; P 5 0.016).

DISCUSSION COPD is associated with inflammation and ineffective repair of the injured epithelium and loss of structural integrity. Recently, we have shown that these changes may result from dysregulated efferocytosis (defective clearance of apoptotic bronchial epithelial cells) by AMs in the airways (2–5). The resultant net increase in apoptotic material in COPD may perpetuate the airway inflammation via secondary necrosis of this material; therefore, novel treatments that focus on improving macrophage function or reducing apoptosis of bronchial epithelial cells may have clinical significance as new or adjunct treatments for COPD. In this regard, we have previously reported that azithromycin, used at low doses, improved AM phagocytic function in vitro (9). The mechanisms for the azithromycinmediated effects have not been determined.

Lung collectins (primarily SP-A, SP-D, and MBL) have important roles in mediating host defense by facilitating AM phagocytosis of pathogens (10, 11). These molecules also regulate the clearance of apoptotic cells via receptors that bind to their collagenous tail (12). The relevance of the collectin system with regard to the pathogenesis of COPD has therefore been the subject of recent interest. Certain polymorphisms of MBL have been associated with increased infection risk in COPD (23), but the association with COPD per se has not been extensively studied. Smoking leads to reduced levels of SP-A and SP-D (24, 25). The relevance of perturbations in the collectin pathway has been illustrated in a number of transgenic mouse studies. MBLdeficient mice demonstrate increased susceptibility to a range of infections, and SP-A–deficient mice are also susceptible and have dysregulated inflammatory responses (i.e., proinflammatory mediators are high) (26). SP-D–deficient mice spontaneously develop emphysema and subpleural fibrosis. Vandivier and colleagues showed that the clearance of intrapulmonary instilled apoptotic cells was reduced in the SP-D–deficient mice (12). We therefore hypothesized that COPD would be associated with perturbations in the collectin system and that the prophagocytic effects of azithromycin could be at least partially linked to changes in this system. We found that levels of soluble collectins, MBL, and SP-D were significantly reduced in BAL from subjects with COPD compared with healthy control subjects. Consistent with our findings of reduced expression of other recognition molecules (5), AMs from current- and ex-smoker patients with COPD were also shown to express significantly less MR than AMs from healthy control subjects. A previous study (25) reported that the

Figure 4. Effect of azithromycin treatment on apoptosis of epithelial cells collected by bronchial brushing. (A) Reduction in the % apoptosis of bronchial epithelial cells after treatment of subjects with chronic obstructive pulmonary disease (COPD) with low-dose azithromycin for 12 weeks. Broken lines and solid lines represent current-smoker and ex-smoker subjects with COPD, respectively. (B) Data summarized in box plots as described in Figure 2. *Significant decrease in epithelial cell apoptosis after azithromycin treatment compared with before treatment. Apoptosis was assessed using 7amino-actinomycin D (7-AAD).


145

Figure 5. Effect of azithromycin treatment on expression of mannose receptor (MR) by alveolar macrophages. (A) Significant increase in MR after treatment of subjects with mannose chronic obstructive pulmonary disease (COPD) with low-dose azithromycin for 12 weeks. Broken lines and solid lines represent current-smoker and exsmoker subjects with COPD, respectively. (B) Data summarized in box plots as described in Figure 2. *Significant increase in the mean fluorescence intensity (MFI) staining for MR after azithromycin treatment compared with before treatment. There were no significant changes in levels of mannose-binding lectin or surfactant protein-D after azithromycin therapy.

levels of SP-D in BAL decreased with smoking but were not influenced by age. Consistent with this study, we found no correlation between age and SP-D or MR; however, we noted a weak but significant negative correlation between MBL and age. The previous study (25) noted no difference in SP-D levels between smokers with evidence of emphysema on computed tomography scan and those who had normal CT scans. However, they did not assess ex-smokers with emphysema. Our study shows a sustained reduction in SP-D in COPD even after smoking cessation. In our in vitro experiments, low-dose azithromycin significantly increased the expression of MR by 50%. A functional link between the reduced expression of mannose receptor and

phagocytosis in COPD was established by blocking the expression of MR on AMs using a specific blocking antibody. We found that this significantly reduced the phagocytic ability of AMs by 60%. To our knowledge, this is the first report of this technique being applied with regard to AM phagocytosis of apoptotic cells. Taken together, these findings strongly implicate the mannose receptor in the defective phagocytic function of AM in COPD and as a target for the azithromycin-mediated improvement in phagocytic ability in vivo. To further address this issue, we administered azithromycin at low dose to a cohort of subjects with COPD. The study was open-labeled and uncontrolled and primarily focused on objective biological responses obtained from the bronchoscopy

Figure 6. Time course of changes in (A) Total leukocyte count (WCC) and (B) C-reactive protein (CRP) in peripheral blood after azithromycin treatment in patients with chronic obstructive pulmonary disease. Data are presented as mean 1 SEM. *Significant decrease in total WCC or CRP after 12 weeks of treatment with low-dose azithromycin.

146


TABLE 2. EFFECT OF AZITHROMYCIN ON INFLAMMATORY CYTOKINE LEVELS IN BLOOD, BRONCHOALVEOLAR LAVAGE, AND BREATH CONDENSATE*

Plasma Pre-azi Post-azi BAL Pre-azi Post-azi Breath Pre-azi Post-azi

TNF-a

IL-8

IL-6

IL-1b

5.6 6 0.99 3.5 6 0.54

22.1 6 3.3 18.9 6 2.7

5.3 6 1.1 5.1 6 1.2

68.6 6 23.5 53.5 6 27.4

5.9 6 1.7 4.4 6 0.7

992 6 524 547 6 310

19.3 6 9.5 9.9 6 6.1

47.5 6 10.0 28.5 6 4.7†

3.2 6 0.5 3.1 6 0.3

11.7 6 7.8 5.8 6 0.8

3.0 6 0.5 3.9 6 0.5

14.6 6 2.2 28.6 6 5.0

Definition of abbreviations: BAL 5 bronchoalveolar lavage; Post-azi 5 after administration of azithromycin; Pre-azi 5 before administration of azithromycin; TNF 5 tumor necrosis factor. * Patients with chronic obstructive pulmonary disease were treated with lowdose azithromycin for 12 weeks. cytometric bead array was used to investigate cytokine levels in plasma, BAL, and breath condensate before (Pre-Azi) and after (Post-Azi) administration of azithromycin. † Significant decrease in cytokine expression after azithromycin treatment.

samples taken. For logistics and patient acceptability reasons, we did not perform a crossover format because this would require three bronchoscopy procedures. These subjects were followed for an additional 12 weeks after the treatment phase and second bronchoscopy, during which time lung function and parameters of inflammation from peripheral blood and breath condensate samples were assessed. Consistent with our in vitro findings, we observed improved ability of AMs to phagocytose apoptotic bronchial epithelial cells and increased AM expression of MR. Other molecules that are important in AM recognition of apoptotic cells (CD91, CD31, CD36, and CD44) were not significantly changed by azithromycin therapy. We also noted significantly reduced levels of bronchial epithelial cell apoptosis in the airway after 12 weeks of treatment with low-dose azithromycin. The reasons for the decreased levels of apoptosis in bronchial epithelial cells after azithromycin therapy are probably multifactorial but may be a direct result of the improved phagocytic function of AMs, possibly linked to the increased expression of MR on these cells. Oxidative stress has been reported to induce epithelial cell apoptosis; however, we did not find any significant reduction in markers of oxidative stress after azithromycin therapy. In contrast to our previous study, we did not observe a significant correlation between the macrophage phagocytosis and the extent of bronchial epithelial cell apoptosis, although

this was likely a result of the small patient numbers in the present study. The reduction in systemic inflammation after azithromycin treatment, evidenced by significantly reduced CRP and WCCs and a trend for reduced inflammatory cytokine production in blood, BAL, and breath condensate is consistent with our in vitro findings and reports by others (9). In our previous study, we reported that culturing AMs with low-dose azithromycin reduced the expression of IL-8, IL-6, and TNF-a (9). We noted significant correlations between blood WCC and AM phagocytosis and bronchial epithelial cell apoptosis, suggesting a possible link between systemic inflammation and improved efferocytosis. A larger, blinded, placebo-controlled study is warranted to more thoroughly investigate any links between the changes in biological parameters and clinical outcomes including quality of life and lung inflammation. In contrast to the changes in AMs, no changes were noted for other cell types, including lymphocytes, in response to azithromycin therapy (including T-cell activation, Treg, and cytotoxic T-cell numbers or Th1 bias). This is not surprising given the propensity for azithromycin to reach very high concentrations in AM and the lack of effect of azithromycin that we have previously observed for T cells in vitro (unpublished data). The dose of azithromycin used in the present study was based on our in vitro findings and on a number of literature reports on the use of macrolides in Japanese panbronchiolitis and CF (7, 8, 27). We opted for a dose at the lower end of the ranges used. Kobayashi and colleagues, using azithromycin at 250 mg twice weekly in diffuse panbronchiolitis, noted some degree of clinical efficacy in 44 of 52 patients treated; adverse events were minor (6). Baumann and colleagues analyzed levels of azithromycin detectable in sputum using a 250-mg twice-weekly regimen (7). The interquartile range of concentrations was 0.2 to 1.4 mg/ml, with the range over the 12 weeks noted to be between undetectable, and 5.2 mg/ml. These values generally meet or exceed those for which we have seen effects in vitro. Baumann and colleagues reported that seven of eight subjects on this dose had significant effects on sputum viscoelasticity after 12 weeks of therapy, and the effects were similar to a 250-mg daily regimen, although the low-dose group had a 5-day ‘‘loading’’ dose of 250 mg once daily, which we replicated in our study (7). The results of this study suggest that the long-term use of low-dose azithromycin is an attractive proposition for COPD and for several other lung diseases for which alternate therapies are unavailable or inadequate. The effects of long-term, lowdose azithromycin have not been evaluated with regard to

TABLE 3. PERCENTAGE OF T CELLS EXPRESSING MARKERS OF ACTIVATION (CD25), TH1 TYPE (62L2CD45RO1), OR TREGS (CD127DIMCD25BRIGHT)

Blood Pre-azi Post-azi BAL Pre-azi Post-azi Brushing intraepithelial T cells Pre-azi Post-azi

2008

CD3162L245RO1 (Th1 T cells)

CD31CD81281 (cytotoxic T cells)

CD31CD251 (activated T cells)

CD31CD251CD1272 (Tregs)

25.4 6 2.4 25.2 6 2.3

13.3 6 3.2 13.6 6 2.8

14.3 6 2.5 15.3 6 2.4

4.0 6 0.4 4.4 6 0.6

81.1 6 3.6 76.8 6 2.1

18.7 6 6.1 12.8 6 3.4

17.8 6 1.7 16.0 6 2.9

7.9 6 2.1 7.8 6 2.1

62.2 6 19.2 57.8 6 21.1

21.4 6 12.0 11.2 6 2.9

12.4 6 3.2 11.9 6 2.7

3.2 6 1.0 4.2 6 1.2

Definition of abbreviations: BAL 5 bronchoalveolar lavage; Post-azi 5 after administration of azithromycin; Pre-azi 5 before administration of azithromycin; Tregs 5 regulatory T cells.


COPD exacerbations; however, a recent study (28) applied erythromycin (although not at low dose) for 1 year and noted a reduction in the rate of exacerbations in subjects with COPD. If macrolide antibiotics can reduce exacerbations, it is plausible that one way that they do this is by promoting phagocytosis, although this may have more to do with the phagocytosis of bacteria than efferocytosis of apoptotic epithelial cells. This subject requires further investigation. With regard to long-term, low-dose macrolide therapy, the emergence of resistant strains of bacteria does need to be considered. One CF study published recently looked specifically at this issue (29). The investigators used a higher dose of azithromycin (up to 500 mg three times per week) for up to 720 days. In absolute terms, the isolates of H. influenzae and Staphylococcus sp in sputum fell, but the percentage of those resistant to azithromycin increased. There were no data to indicate that the resistance patterns had any clinical impact. Other studies using long-term, low-dose macrolides in panbronchiolitis or CF have not noted adverse clinical consequences due to changing resistance patterns, but this area requires more research. In our study, no significant colonization of the airways was observed, with all BAL samples showing no growth or a scant growth of oral flora. We acknowledge that the numbers assessed were small and treatment time relatively short. Notwithstanding the importance of this issue, in practical clinical terms there are a number of alternative antibiotics that can be used if infection with an azithromycin-resistant H. infuenzae or Staphylococcus sp should become a clinically relevant concern. In conclusion, this study provides important new insights into the biological basis of macrophage dysfunction in COPD. Our findings of improved efferocytosis and reduced systemic inflammation after medium-term, low-dose azithromycin therapy indicate a novel approach to address the perturbations we have found and provide further rationale to investigate macrolides as supplements to existing therapies in COPD. Treatment of patients who have COPD in this way may improve clearance of accumulated apoptotic material and reduce the risk of secondary necrosis and release of toxic cell contents that perpetuate inflammation. Conflict of Interest Statement: S.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. G.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. H.J. has been a speaker at medical meetings financed by various pharmaceutical companies (GlaxoSmithKline, AstraZeneca, Pfizer, Boehringer Ingelheim) on the topics of asthma care and COPD (total payments received over the past 3 years AU$23,000). G.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. J.A. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. M.H. has been paid lecture fees by GlaxoSmithKline ($2,000 in 2006, 2007), AstraZeneca ($3,000 in 2006, 2007), and Pfizer ($3,000 in 2006, 2007). P.N.R. received an unrestricted grant from Pfizer Pharmaceuticals of $50,000, which partly supported this project. Acknowledgment: The authors acknowledge the invaluable contribution of the Bronchoscopy Unit and the nursing staff at the Chest Clinic of the Thoracic Medicine Department, Royal Adelaide Hospital, Adelaide, Australia.

References 1. World Health Organization. The World Health Report 2002. Reducing risks, promoting healthy life. Geneva: World Health Organization; 2002. 2. Hodge S, Hodge G, Scicchitano R, Reynolds PN, Holmes M. Alveolar macrophages from subjects with chronic obstructive pulmonary disease are deficient in their ability to phagocytose apoptotic airway epithelial cells. Immunol Cell Biol 2003;81:289–296. 3. Hodge S, Hodge G, Holmes M, Reynolds PN. Increased airway epithelial and T-cell apoptosis in COPD remains despite smoking cessation. Eur Respir J 2005;25:447–454.

147

4. Hodge SJ, Hodge GL, Reynolds PN, Scicchitano R, Holmes M. Increased production of TGF-beta and apoptosis of T lymphocytes isolated from peripheral blood in COPD. Am J Physiol Lung Cell Mol Physiol 2003;285:L492–L499. 5. Hodge S, Hodge G, Ahern J, Jersmann H, Holmes M, Reynolds PN. Smoking alters alveolar macrophage recognition and phagocytic ability: implications in COPD. Am J Respir Cell Mol Biol 2007;37: 748–755. 6. Kobayashi H, Takeda H, Sakayori S, Kawakami Y, Otsuka Y, Tamura M, Konishi K, Tanimoto S, Fukakusa M, Shimada K. Study on azithromycin in treatment of diffuse panbronchiolitis [in Japanese]. Kansenshogaku Zasshi 1995;69:711–722. 7. Baumann U, King M, App EM, Tai S, Konig A, Fischer JJ, Zimmermann T, SextroW, von der Hardt H. Long term azithromycin therapy in cystic fibrosis patients: a study on drug levels and sputum properties. Can Respir J 2004;11:151–155. 8. Patel KB, Xuan D, Tessier PR, Russomanno JH, Quintiliani R, Nightingale CH. Comparison of bronchopulmonary pharmacokinetics of clarithromycin and azithromycin. Antimicrob Agents Chemother 1996;40:2375–2379. 9. Hodge S, Hodge G, Brozyna S, Jersmann H, Holmes M, Reynolds PN. Azithromycin increases phagocytosis of apoptotic bronchial epithelial cells by alveolar macrophages. Eur Respir J 2006;28:486–495. 10. Stuart LM, Henson PM, Vandivier RW. Collectins: opsonins for apoptotic cells and regulators of inflammation. Curr Dir Autoimmun 2006;9:143–161. 11. van de Wetering JK, van Golde LMG, Batenburg JJ. Collectins: players of the innate immune system. Eur J Biochem 2004;271:1229–1249. 12. Vandivier RW, Ogden CA, Fadok VA, Hoffmann PR, Brown KK, Botto M, Walport MJ, Fisher JH, Henson PM, Greene KE. Role of surfactant proteins A, D, and C1q in the clearance of apoptotic cells in vivo and in vitro: calreticulin and CD91 as a common collectin receptor complex. J Immunol 2002;169:3978–3986. 13. Hodge S, Ahern J, Matthews G, Hodge G, Jersmann H, Holmes M, Reynolds PN. Low dose azithromycin improves efferocytosis (apoptosis and macrophage function) and reduces inflammation in COPD patients. Respirology 2007;12:12. 14. Wileman T, Boshans R, Stahl P. Uptake and transport of mannosylated ligands by alveolar macrophages: studies on ATP-dependent receptor-ligand dissociation. J Biol Chem 1985;260:7387–7393. 15. Stahl PD, Wileman TE, Diment S, Shepherd VL. Mannose-specific oligosaccharide recognition by mononuclear phagocytes. Biol Cell 1984;51:215–218. 16. Beharka AA, Gaynor CD, Kang BK, Voelker DR, McCormack FX, Schlesinger LS. Pulmonary surfactant protein A up-regulates activity of the mannose receptor, a pattern recognition receptor expressed on human macrophages. J Immunol 2002;169:3565–3573. 17. Hodge G, Osborn M, Hodge S, Nairn J, Tapp H, Kirby M, Sepulveda H, Morgan E, Revesz T, Zola H. Rapid simultaneous measurement of multiple cytokines in childhood oncology patients with febrile neutropenia: increased interleukin (IL)-8 or IL-5 correlates with culturepositive infection. Br J Haematol 2006;132:247–248. 18. Hodge G, Hodge S, Reynolds PN, Holmes M. Compartmentalization of intracellular proinflammatory cytokines in bronchial intraepithelial T cells of stable lung transplant patients. Clin Exp Immunol 2006;145: 413–419. 19. Seddiki N, Santner-Nanan B, Martinson J, Zaunders J, Sasson S, Landay A, Solomon M, Selby W, Alexander SI, Nanan R, et al. Expression of interleukin (IL)-2 and IL-7 receptors discriminates between human regulatory and activated T cells. J Exp Med 2006;203:1693–1700. 20. Matsuzaki S, Shinozaki K, Kobayashi N, Agematsu K. Polarization of Th1/Th2 in human CD4 T cells separated by CD62L: analysis by transcription factors. Allergy 2005;60:780–787. 21. Matthews GM, Howarth GS, Butler RN. Short-chain fatty acid modulation of apoptosis in the Kato III human gastric carcinoma cell line. Cancer Biol Ther 2007;6:1051–1057. 22. Jackson AS, Sandrini A, Campbell C, Chow S, Thomas PS, Yates DH. Comparison of biomarkers in exhaled breath condensate and bronchoalveolar lavage. Am J Respir Crit Care Med 2007;175:222–227. 23. Yang IA, Seeney SL, Wolter JM, Anders EM, McCormack JG, Tunnicliffe AM, Rabnott GC, Shaw JG, Dent AG, Kim ST, et al. Mannose-binding lectin gene polymorphism predicts hospital admissions for COPD infections. Genes Immun 2003;4:269–274. 24. Honda Y, Takahashi H, Kuroki Y, Akino T, Abe S. Decreased contents of surfactant proteins A and D in BAL fluids of healthy smokers. Chest 1996;109:1006–1009.

148


25. Betsuyaku T, Kuroki Y, Nagai K, Nasuhara Y, Nishimura M. Effects of ageing and smoking on SP-A and SP-D levels in bronchoalveolar lavage fluid. Eur Respir J 2004;24:964–970. 26. Shi L, Takahashi K, Dundee J, Shahroor-Karni S, Thiel S, Jensenius JC, Gad F, Hamblin MR, Sastry KN, Ezekowitz RA. Mannose-binding lectin-deficient mice are susceptible to infection with Staphylococcus aureus. J Exp Med 2004;199:1379–1390. 27. Kadota J, Mukae H, Mizunoe S, Kishi K, Tokimatsu I, Nagai H, Tomono K, Kohno S, Nasu M. Long-term macrolide antibiotic

2008

therapy in the treatment of chronic small airway disease clinically mimicking diffuse panbronchiolitis. Intern Med 2005;44:200–206. 28. Seemungal TAR, Wilkinson TMA, Perera W, Hurst J, Wedzicha JA. Long term macrolide therapy decreases exacerbations in COPD. Am J Respir Crit Care Med 2007;175:A764. 29. Phaff SJ, Tiddens HA, Verbrugh HA, Ott A. Macrolide resistance of Staphylococcus aureus and Haemophilus species associated with longterm azithromycin use in cystic fibrosis. J Antimicrob Chemother 2006;57:741–746.