Glucocorticoid sensitivity of lipopolysaccharide-stimulated chronic obstructive pulmonary disease alveolar macrophages

Abstract

It has been reported that alveolar macrophages from patients with chronic obstructive pulmonary disease (COPD) display glucocorticoid (Gc) resistance. The Gc sensitivity of inflammatory mediators released by COPD macrophages may vary. The objective of this study was to identify Gc-insensitive inflammatory mediators produced by lipopolysaccharide (LPS)-stimulated alveolar macrophages from COPD patients. LPS-stimulated alveolar macrophages from 15 COPD patients, nine smokers (S) and nine healthy non-smokers (HNS) were stimulated with LPS with or without dexamethasone (100 and 1000 nM). Luminex and enzyme-linked immunosorbent assay were used to measure 23 inflammatory mediators. After LPS stimulation there were lower levels of inflammatory mediators in COPD patients and S compared to HNS. There was no difference between groups for the effects of dexamethasone at either concentration (P > 0·05 for all comparisons). Tumour necrosis factor (TNF)-α, interleukin (IL)-6 and growth-related oncogene (GRO)-α displayed the greatest sensitivity to dexamethasone in COPD patients, while IL-8, granulocyte–macrophage colony-stimulating factor (GM-CSF) and granulocyte colony-stimulating factor (G-CSF) were the least sensitive. COPD macrophages have a reduced response to LPS. Gc sensitivity was similar in COPD macrophages compared to controls. We identify some Gc-insensitive cytokines, including GM-CSF, G-CSF and IL-8, that may be involved in the progression of airway inflammation in COPD patients.

Introduction

Chronic obstructive pulmonary disease (COPD) is characterized by progressive airway inflammation [1]. Glucocorticoids (Gc) are the most widely used anti-inflammatory drugs in COPD, suppressing the activity of key transcription factors such as nuclear factor kappa B (NFκB) leading to decreased inflammatory gene expression [2]. However, the clinical benefits of Gc in COPD patients are limited [3]. This phenomenon has been called Gc resistance [4].

The number of macrophages are increased in the lungs of COPD patients [5,6]. These cells release a range of proinflammatory mediators, and are thought to play a key role in the pathogenesis of COPD. It has been reported that the inhibitory effects of Gc on cytokine production from COPD alveolar macrophages cultured ex-vivo are reduced compared to controls [7,8]. Various molecular mechanisms have been proposed to explain this apparent decreased sensitivity of COPD alveolar macrophages to Gc, including oxidative stress-induced remodelling of the chromatin structure of inflammatory genes and reduced histone deacetylase activity [9,10].

Using multiplex protein profiling, it has been shown that the effects of Gc vary between different cytokines produced by lipopolysaccharide (LPS)-stimulated alveolar macrophages [11] and peripheral blood mononuclear cells [12] from patients with severe asthma. A similar phenomenon has also been observed using gene arrays in monocyte-derived macrophages from patients with COPD [13]. Previous studies of the effects of Gc on cytokine production from COPD alveolar macrophages cultured ex-vivo have investigated a limited number of cytokines, namely interleukin (IL)-8, granulocyte–macrophage colony-stimulating factor (GM-CSF) and tumour necrosis factor (TNF)-α[7,8]. It would be important to profile a wider range of inflammatory mediators produced by COPD alveolar macrophages; Gc-insensitive inflammatory mediators may represent potential therapeutic targets. The primary aim of this study was to identify the most Gc-resistant inflammatory mediators produced by COPD alveolar macrophages by profiling a range of cytokines and chemokines. We have studied the pharmacological effects of the Gc dexamethasone on LPS-stimulated alveolar macrophages from COPD patients, and control groups of smokers and healthy non-smokers.

Methods

Study subjects

Fifteen COPD patients, diagnosed in accordance with current Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines [14], nine smokers (S) with no airway obstruction and nine healthy non-smokers (HNS) were recruited for bronchoscopy (see Table 1 for demography). Eight COPD patients were on inhaled corticosteroids (ICS); nine COPD patients were current smokers (Table 1). For experiments to optimize the cell culture conditions, four COPD patients undergoing lung surgery resection were also recruited: three were males with mean age 62·5 years, mean forced expiratory volume in 1 s (FEV₁)% predicted of 69 and a mean pack year history of 54. All were current smokers and two were using inhaled corticosteroids. All subjects gave written informed consent. The study was approved by the local research ethics committee.

Table 1.

Bronchoalveolar lavage (BAL) subject demography.

	HNS	S	COPD
Sex (male/female)	3/6	1/8	13/2
Age (years)	60 (11·54)	56 (3·82)	64·3 (4·08)
FEV1	2·88 (0·59)	2·77 (0·77)	1·91 (0·37)
FEV1 % predicted	111·89 (18·4)	110·78 (9·34)	63·53 (8·68)
FEV1/FVC	71·9 (5·3)	73·4 (4·4)	48·3 (7·4)
Pack year history	0	33·1 (18·5–55)	60·5 (31–106)
Inhaled corticosteroid users	0	0	8
Beclamethasone dipropionate equivalent (BDP)	0	0	1500 (480–2000)
Current smokers	0	9	9

BAL subject demography: spirometry, smoking history and corticosteroid use of HNS, S and chronic obstructive pulmonary disease (COPD) groups. Normal data are presented as mean ± standard deviation and non-parametric as median (range) and were compared between all subject groups using analysis of variance (anova). FEV₁: forced expiratory volume in 1 s; FVC: forced vital capacity.

Study design

Bronchoscopy

Bronchoscopies were performed after the patients had been sedated. The bronchoscope was wedged peripherally and 0·9% (wt/vol) warmed normal saline instilled in four 60 ml aliquots into the bronchial tree. This procedure was repeated in the other lung, to give a total volume instilled of 480 ml. Retrieved bronchoalveolar lavage (BAL) fluid was placed on ice.

Alveolar macrophage isolation

BAL fluid was filtered (100 µm filter) and centrifuged (400 g for 10 min) to obtain a cell pellet. Cells were resuspended in RPMI-1640 (Sigma-Aldrich, Poole, Dorset, UK) supplemented with 10% (v/v) fetal calf serum (FCS; Invitrogen, Paisley, Scotland, UK), 2 mM L-glutamine (Invitrogen), 100 U/ml penicillin and 100 µg/ml streptomycin (Sigma-Aldrich). Cell viability was assessed by trypan blue exclusion and cytospins were prepared by cytocentrifugation at 7000 g, air-drying for 30 min, followed by methanol fixation and staining with RAPI-DIFF II (Triangle Biomedical Science, Durham, NC, USA). BAL alveolar macrophages were isolated by plastic adhesion at a cell density of 1 × 10⁶/ml (100 000 cells were seeded into each well of 96-well plates) for 2 h in a 5% CO₂ humidified atmosphere at 37°C. The isolated macrophages were washed with pre-warmed supplemented RPMI-1640 media removing non-adherent cells.

For resected lung tissue, areas far distant from the tumour were perfused with 0·1 M NaCl to isolate macrophages. Retrieved fluid was centrifuged (10 min, 400 g, room temperature), the cell pellet resuspended in RPMI-1640 and the cell suspension floated over a Ficol gradient, centrifuged and cells counted by trypan blue exclusion. Cells were resuspended in RPMI-1640 and treated as described for alveolar macrophages.

Cell culture

In preliminary experiments, lung resected macrophages (n = 4) were stimulated with and without LPS (0·01–10 µg/ml, serotype O26:B6) for 2, 4, 6, 8, 12, 24 and 48 h to select a time-point and concentration of LPS for future experiments involving alveolar macrophages. The full data for these optimization experiments are shown in the Results section. BAL alveolar macrophages were stimulated with and without LPS (0·01–10 µg/ml) to confirm the submaximal concentration of LPS to be used to study dexamethasone. Alveolar macrophages were incubated with or without dexamethasone [reconstituted with dimethyl sulphoxide (DMSO) and diluted in supplemented RPMI-1640] for 1 h followed by 1 µg/ml LPS stimulation (serotype O26:B6, in supplemented RPMI-1640) for 4 h. Experiments performed with dexamethasone but without LPS stimulation resulted in cytokine levels often being below the lower limit of detection, making statistical analysis impractical; these data are not shown in this paper. Cell culture supernatants were stored at −80°C prior to analysis.

Cytokine and chemokine assays

A Luminex kit (Millipore, Massachusetts, USA) was used according to the manufacturer's instructions to measure the following cytokines: IL-1β, IL-2, IL-4, IL-5, IL-6, IL-7, IL-10, IL-12, IL-13, IL-15, IL-17, interferon (IFN)-γ, granulocyte colony-stimulating factor (G-CSF), GM-CSF, TNF-α, Eotaxin, monocyte chemoattractant protein (MCP)-1, interferon-inducible protein (IP)-10 and regulated upon activation normal T cell expressed and secreted (RANTES). The lower limit of quantification was 3·2 pg/ml. Where stated, the following cytokines were also measured by enzyme-linked immunosorbent assay (ELISA) according to the manufacturer's instructions (R&D Systems, Abingdon, UK) to quantitate growth-related oncogene (GRO)-α, IL-8, macrophage inflammatory protein (MIP)-1α, IL-6 and TNF-α. The lower limits of quantification were 31·25 pg/ml for GRO-α and IL-8, 7·81 pg/ml for MIP-1α, 4·69 pg/ml for IL-6 and 15·625 pg/ml for TNF-α.

RNA preparation

Samples were homogenized by drawing three times through a 25-gauge needle to disrupt genomic DNA by mechanical shearing. Phase separation was achieved following centrifugation at 12 000 g for 15 min of 0·2 ml chloroform per ml of TRIzol used. The upper aqueous RNA phase was transferred to a fresh tube and purified using RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacture's instructions and incorporating the DNase digestion step. mRNA purity and concentration were assessed on the NanoDrop ND1000 spectrophotometer to establish the 260/280 ratio.

Real-time quantitative–polymerase chain reaction (RT–PCR)

It was decided a priori to analyse the levels of IL-8, TNF-α, MIP-1α, IL-6 and GRO-α. Primers and TaqMan probes for these inflammatory mediators were designed using Primer Express version 2·0 (Applied Biosystems, California, USA) and supplied by Eurogentec Ltd (Seraing, Belgium). Sequences were as follows:

IL-8:	3′ CTGGCCGTGGCTCTCTTG, 5′ TTAGCACT CCTTGGCAAAACTG
Taq P:	ACCTTCACACAGAGCTGCAGAAATCAGGA AG
TNF-α:	3′ CCCAGGCAGTCAGATCATCTTC, 5′ AGCTGCCCCTCAGCTTGA
Taq P:	CAAGCCTGTAGCCCATGTTGTAGCAAACC
MIP-1α:	3′ TGGCTCTCTGCAACCAGTTCT, 5′ GCCGGGAGGTGTAGCTGAA
Taq P:	TGACACGCCGACCGCCTGC
IL-6:	3′ GGTACATCCTCGACGGCATCT, 5′ GTGCCTCTTTGCTGCTTTCAC
Taq P:	TGTTACTCTTGTTACATGTCTCCTTTCTCAG GGCT
GRO-α	3′ ACATGCCAGCCACTGTGATAGA, 5′ TTCCCCTGCCTCACAATG
Taq P:	CGGATCCAAGCAAATGGCCAATGA

The probes were labelled with 6-carboxyfluorescein (FAM) at their 5′-terminal end and were quenched with 6-carboxytetramethylrhodamine (TAMRA) on their 3′ terminal and were optimized before use.

TaqMan PCR assays were performed on a Stratagene Mx3000 (Stratagene, La Jolla, CA, USA) as a one-step RT–PCR using 0·1 µg of RNA per mediator. Standard curves were run on the same plate using serial dilutions of universal human lung reference RNA (Stratagene). Relative gene expression was determined by comparison using the standard curve. Total RNA was measured using Ribogreen (Invitrogen) and used to normalize for the quantity of RNA used for PCR.

Data analysis

Normality was assessed using the Kolmogorov–Smirnov test. Data were compared between subject groups using analysis of variance (anova) for parametric data and using the Kruskal–Wallis test for non-parametric data. If P < 0·05 by anova, then unpaired Student's t-tests for parametric data and Mann–Whitney U-tests for non-parametric data were used for pairwise comparisons. The infectious dose 50% (IC₅₀) was calculated from the group mean data at each concentration of dexamethasone. The effect of smoking and ICS use on cytokine responses and dexamethasone effects were evaluated using unpaired t-tests. All statistical analysis was performed using Graphpad InStat version 3·00 for Windows 95 (GraphPad Software, San Diego, CA, USA; http://www.graphpad.com).

Results

In two COPD patients there was no BAL fluid recoverable, so only 13 COPD samples were available for analysis. The majority of these 13 COPD patients were GOLD stage II (n = 12), with one GOLD stage III patient. The gender of the patients was not matched between groups, with more females present in the S group. The BAL volume recovered was significantly lower in COPD patients compared to S and HNS (see Table 2, P < 0·0001 by anova). The absolute number of macrophages recovered was highest in S and lowest in HNS (P < 0·0001 by anova). Similarly, the number of macrophages/ml of BAL was greatest in S and lowest in HNS (P < 0·0001).

Table 2.

Cell counts in bronchoalveolar lavage.

	HNS	S	COPD	anova
No.	9	9	13
Volume instilled (ml)	467·5 ± 66·4	446·7 ± 60·8	465·5 ± 74·1	0·7501
Volume recovered (ml)	212·5 ± 70·9	168·7 ± 43	93·3 ± 50·7	< 0·0001
Recovery, %	45·1 ± 12·7	37·6 ± 7·9	19·9 ± 14·1	0·0002
Total absolute macrophage count	15·2 ± 7·4	90·4 ± 44·5	18·7 ± 17·2	< 0·0001
Total alveolar macrophage count, 106 cells/ml	0·07 ± 0·04	0·54 ± 0·22	0·18 ± 0·13	< 0·0001
Bronchial epithelial cells, %	4·5 (3·8–12·8)	12·0 (4·5–13·3)	12·5 (5·0–15·0)	0·7328
Macrophages, %†	88·8 (85·5–90·7)	89·0 (85·3–92·4)	94·5 (89·7–95·4)	0·6350
Neutrophils, %†	0·4 (0–1·9)	1·4 (0·5–2·0)	0·3 (0–1·0)	0·7623
Lymphocytes, %†	10·5 (7·7–12·2)	9·5 (7·4–11·6)	5·2 (4·8–6·0)	0·0075

^†Percentage of total non-bronchial epithelial cells. Bronchoalveolar lavage fluid (BALF) alveolar macrophage cell counts and differential counts. Normal data are presented as mean ± standard deviation and non-parametric as median (interquartile range) and were compared between all subject groups using analysis of variance (anova). Total absolute macrophage counts are the total number of macrophages in the BALF presented as (×10⁶) cells. Total alveolar macrophage counts are the total number of macrophages in the BALF (×10⁶) divided by volume of BALF recovered (ml), presented as (×10⁶ cells per ml). COPD: chronic obstructive pulmonary disease; HNS: healthy non-smokers; S: smokers.

Optimization of LPS stimulation

In preliminary experiments using lung resection alveolar macrophages, LPS caused a significant induction of TNF-α production (measured by ELISA) by 4 h, which reached a plateau at 6–24 h (Fig. 1). Most of the levels of TNF-α in time-matched controls were lower than the level of detection of the assay. Four h was chosen as a time-point for future experiments as protein production was only marginally submaximal to 6 h, and this would be a suitable time-point to also measure mRNA levels. One µg/ml LPS was shown to be a suitable concentration that causes near maximal stimulation in both lung resection and alveolar macrophages (Fig. 2), and this concentration was used for further experiments with dexamethasone.

Fig. 1.

Time-course of lipopolysaccharide (LPS)-stimulated tumour necrosis factor (TNF)-α production from chronic obstructive pulmonary disease patients’ alveolar macrophages. Concentrations of TNF-α measured in alveolar macrophage (n = 4) cell culture supernatants show the effect of 1 and 10 µg/ml LPS stimulation. Non-parametric data shown as median and interquartile range. ***P < 0·001, **P < 0·01 and *P < 0·05 compared to unstimulated time-matched controls, of which most were lower than the level of detection of assay. TNF-α measured by enzyme-linked immunosorbent assay.

Fig. 2.

Dose–response of lipopolysaccharide (LPS)-stimulated tumour necrosis factor (TNF)-α production from chronic obstructive pulmonary disease (COPD) patients’ alveolar macrophages. Concentrations of TNF-α measured in alveolar macrophage cell culture supernatants show the effect of 0·01–10 µg/ml LPS stimulation. Alveolar macrophages from healthy non-smoker (HNS), smoker (S) and COPD groups are shown. Mean ± SEM shown. ***P < 0·001, **P < 0·01 and *P < 0·05 compared to unstimulated levels. TNF-α measured by enzyme-linked immunosorbent assay.

LPS-stimulated inflammatory mediator production

Using Luminex, IL-1β, IL-2, IL-3, IL-4, IL-5, IL-7, IL-12, IL-13, IL-15, IL-17, IFN-γ and Eotaxin were below the lower limit of detection with and without stimulation with LPS. Data on the other inflammatory mediators measured by Luminex, as well as GRO-α, MIP-1α and IL-8 measured by ELISA, are presented in Fig. 3.

Fig. 3.

Unstimulated and lipopolysaccharide (LPS, 1 µg/ml)-stimulated inflammatory mediator production from alveolar macrophages. Concentrations of 11 cytokines in (a) unstimulated and (b) 1 mg/ml LPS-stimulated alveolar macrophage cell culture supernatants. Alveolar macrophages from healthy non-smoker (HNS), smoker (S) and chronic obstructive pulmonary disease (COPD) groups are shown. Parametric data set shown as mean ± SEM. Granulocyte colony-stimulating factor (G-CSF) and interferon-inducible protein (IP)-10 were non-parametric data sets shown with median, interquartile range and range, as box-plots. Enzyme-linked immunosorbent assay was used to quantify growth-related oncogene (GRO)-α, interleukin (IL)-8 and macrophage inflammatory protein (MIP)-1α. Luminex was used to measure IL-6, monocyte chemoattractant protein (MCP)-1, tumour necrosis factor (TNF)-α, granulocyte–macrophage colony-stimulating factor (GM-CSF), IL-10, regulated upon activation normal T cell expressed and secreted (RANTES), G-CSF and IP-10. ***P < 0·001, **P < 0·01 and *P < 0·05 between groups.

The levels of inflammatory proteins produced by unstimulated alveolar macrophages after 4 h was different between groups; anova showed significant differences (P < 0·05) between groups for GRO-α, IL-8, MIP-1α, IL-6, TNF-α, GM-CSF, IL-10, RANTES, MCP-1, G-CSF and IP-10 (Fig. 3a). Subsequent two-way tests (unpaired t-tests for all parametric data and Mann–Whitney tests for non-parametric G-CSF and IP-10 data) showed that COPD patients and S produced significantly lower (P < 0·05) levels of GRO-α, IL-8, MIP-1α, IL-6, TNF-α and RANTES compared to HNS. GM-CSF, IL-10, G-CSF and IP-10 production from COPD patients was also significantly lower compared to HNS. There were no differences between COPD patients and S.

After LPS stimulation, there were numerically lower levels of all inflammatory mediators in COPD patients and S compared to HNS, except for MIP-1α (Fig. 3b). Using two-way anova, these differences reached statistical significance (P < 0·05) for GRO-α, IL-8, IL-6 and GM-CSF.

Dexamethasone inhibition of LPS-stimulated cytokine proteins

LPS-stimulated cells were treated with 0·1 µM and 1 µM dexamethasone (Fig. 4). Luminex was performed, as well as ELISAs for GRO-α, MIP-1α and IL-8. There was statistically significant (P < 0·05) inhibition of GRO-α, IL-8, MIP-1α, IL-6, TNF-α, IL-10, RANTES and MCP-1 production in all three subject groups. There was no difference between groups for the effects of dexamethasone at either concentration on these cytokines (P > 0·05 for all comparisons). Both dexamethasone concentrations did not inhibit GM-CSF or G-CSF production in any of the three groups. TNF-α displayed the greatest sensitivity to dexamethasone in COPD patients, while IL-8, GMCSF and GCSF were the least sensitive. The effect of dexamethasone on TNF-α production was measured by both ELISA and Luminex. Extremely similar results were observed with no statistical difference between the methods. The maximal percentage inhibition of 1 µg/ml LPS-stimulated TNF-α production for HNS, S and COPD patients was 76, 83 and 74%, measured by Luminex and 71, 76 and 69%, respectively, measured by ELISA (P > 0·05 for comparisons between methods).

Fig. 4.

Inhibition of inflammatory mediator production by dexamethasone in alveolar macrophages. (a) Per cent inhibition of inflammatory mediators in response to 1 µg/ml lipopolysaccharide (LPS) stimulation by 0·1 µM dexamethasone. (b) Per cent inhibition of inflammatory mediators in response to 1 µg/ml LPS stimulation by 1 µM dexamethasone. Parametric data set shown as mean ± 95% CI. Granulocyte colony-stimulating factor (G-CSF) and interferon-inducible protein (IP)-10 were non-parametric data sets shown with median, interquartile range and range as box-plots. All inhibition shown is statistically significant except for granulocyte–macrophage colony-stimulating factor (GM-CSF) and G-CSF. Enzyme-linked immunosorbent assay was used to quantify growth-related oncogene (GRO)-α, interleukin (IL)-8 and macrophage inflammatory protein (MIP)-1. Luminex was used to measure IL-6, monocyte chemoattractant protein (MCP)-1, tumour necrosis factor (TNF)-α, GM-CSF, IL-10, regulated upon activation normal T cell expressed and secreted (RANTES), G-CSF and IP-10. HNS, healthy non-smokers; S, smokers; COPD, chronic obstructive pulmonary disease.

For patients with sufficient cells remaining (seven COPD patients and nine HS) a dose–response curve to dexamethasone (0·001–10 µM) was constructed for IL-6, IL-8, TNF-α, GRO-α and MIP-1α measured by ELISA (Table 3). There were insufficient cells from HNS for this experiment. IL-8 had the highest IC₅₀ values in both COPD patients and S, while TNF-α and IL-6 had the lowest IC₅₀ values. The effect of a very high concentration of dexamethasone (10 µM) was also different between the cytokines measured (anovaP = 0·0072 for COPD patients and P < 0·0001 for S); again, IL-8 was the most Gc-insensitive cytokine in both COPD patients and S, while TNF-α and IL-6 were the most sensitive. The IC₅₀ values and the effect of dexamethasone 10 µM was similar between COPD patients and S for all cytokines (P > 0·05 for all comparisons).

Table 3.

Infectious dose 50% (IC₅₀) values of dexamethasone from alveolar macrophages (1 µg/ml lipopolysaccharide).

	COPD		S
Mediator	IC50 (nM)	Inhibition at 10 µM (%)	IC50 (nM)	Inhibition at 10 µM (%)
IL-8	524	49·7 ± 15	871	53·7 ± 6·5
GRO-α	162	60·2 ± 13·5	78	69·6 ± 9·2
MIP-1α	93	57 ± 6	81	65·1 ± 4·9
IL-6	85	79·5 ± 12·2	21	90·8 ± 1·7
TNF-α	34	66·5 ± 19·8	37	88·2 ± 7·5

Enzyme-linked immunosorbent assay was used to quantitate IC₅₀ values (nM) from chronic obstructive pulmonary disease (COPD) and smoker (S) groups. The lower limits of quantification were 31·25 pg/ml for growth-related oncogene (GRO)-α and interleukin (IL)-8, 7·81 pg/ml for major inflammatory protein (MIP)-1α, 4·69 pg/ml for IL-6 and 15·625 pg/ml for tumour necrosis factor (TNF)-α.

Effects of ICS and smoking status in COPD patients

ICS use in COPD patients had no effect on the cytokine production in response to LPS; the LPS dose–response curve for TNF-α was similar in current smokers compared to ex-smokers, with no difference between groups in TNF-α levels at each concentration of LPS (see Fig. 5a). Similarly, there were no differences due to ICS use in any of the other cytokines stimulated by LPS (data not shown). The effect of dexamethasone was similar in ICS users and non-users; Fig. 5a shows no difference in the effect of dexamethasone 1 µM on IL-8 and TNF-α production between the groups. Similarly, there were no differences for any of the other cytokines (data not shown).

Fig. 5.

Effects of inhaled corticosteroids (ICS) and smoking status in chronic obstructive pulmonary disease (COPD) patients’ alveolar macrophages. Concentrations of tumour necrosis factor (TNF)-α measured in alveolar macrophage cell culture supernatants show the effect of 0·01–10 µg/ml lipopolysaccharide (LPS) stimulation. TNF-α measured by enzyme-linked immunosorbent assay. Per cent inhibition of inflammatory mediators in response to 1 µg/ml LPS stimulation by 1 µM dexamethasone is also shown for interleukin (IL)-8 and TNF-α, measured by Luminex. Alveolar macrophages from (a) ICS +ve and −ve and (b) current and ex-smoking COPD patients are shown. Mean ± SEM is shown. NS, no significant difference at each concentration of LPS, or with dexamethasone.

Current smoking status had no effect on cytokine production in response to LPS; the LPS dose–response curve for TNF-α was similar in current smokers compared to ex-smokers, with no difference between groups in TNF-α levels at each concentration of LPS (see Fig. 5b). Similarly, there were no differences due to smoking status in any of the other cytokines stimulated by LPS (data not shown). The effect of dexamethasone was similar in current and ex-smokers; Fig. 5b shows no difference in the effect of dexamethasone 1 µM on IL-8 and TNF-α production between the groups. Similarly, there were no differences for any of the other cytokines (data not shown).

Dexamethasone inhibition of mRNA levels

Macrophages from four COPD patients and eight smokers were used to evaluate the effects of dexamethasone on inflammatory gene transcription. There were insufficient cells from the other subjects for these experiments. The RNA was extremely pure, as the 260/280 ratio had a median value of 2·02 (the range for all the samples was 1·72–2·26). LPS up-regulated IL-8, IL-6, TNF-α, GRO-α and MIP-1α mRNA levels at 4 h (Fig. 6); 0·1 µM dexamethasone inhibited transcription significantly for all inflammatory mediators in S. COPD showed inhibition that did not reach significance due to the small sample size. The suppression of TNF-α was greatest, reaching 75% in smokers and 51% in COPD patients. The reduction in IL-8 mRNA levels was only 29% in both groups.

Fig. 6.

Inhibition of inflammatory mediator mRNA production by dexamethasone in alveolar macrophages. Relative mRNA levels of five cytokines were measured in alveolar macrophages from obstructive pulmonary disease (COPD) patients and smokers (S). Data were non-parametric and shown as median ± interquartile range, ***P < 0·001, **P < 0·01 and *P < 0·05 compared to lipopolysaccharide (LPS). GRO, growth-related oncogene; IL, interleukin; MIP, macrophage inflammatory protein; TNF, tumour necrosis factor.

Discussion

The primary aim of this study was to identify Gc-insensitive inflammatory mediators produced by LPS-stimulated COPD alveolar macrophages. Profiling of a range of inflammatory mediators to the effects of two concentrations of dexamethasone (0·1 and 1 µM) revealed some Gc-insensitive cytokines; GM-CSF and G-CSF were not inhibited at all, while IL-8 was suppressed but to a lesser extent than other cytokines such as TNF-α and IL-6. Further confirmation that the effect of dexamethasone varied between cytokines was observed when full dose–response curves were constructed, showing that IL-8 had a far higher IC₅₀ value than TNF-α and IL-6. The effects of dexamethasone were similar in COPD patients and controls. This suggests that the Gc-insensitive production of certain cytokines by alveolar macrophages is an intrinsic property of these cells, even in the healthy state. Previous studies have reported that COPD alveolar macrophages are Gc-insensitive compared to controls [7,8], in contrast to our observations. The reasons for differences between studies will be considered in this discussion.

We used Luminex and ELISA to profile for Gc-insensitive cytokines using two concentrations of dexamethasone (0·1 and 1 µM). The validity of the Luminex results was confirmed when similar data were obtained on the same samples using an ELISA for TNF-α in all three subject groups. We then performed ELISA using macrophages from the subset of patients where there were sufficient cells remaining to construct a full dose–response curve to dexamethasone to assess the suppression of IL-8, IL-6, TNF-α, GRO-α and MIP-1α. These data confirmed the Gc insensitivity of IL-8, as it had the highest IC₅₀ value. In contrast, the IC₅₀ values of TNF-α and IL-6 were much lower.

The effect of dexamethasone on the transcription of IL-8, IL-6, TNF-α, GRO-α and MIP-1α was also assessed; the effect of dexamethasone was again greatest on TNF-α and IL-6. The sample size for this analysis was limited in the COPD group, so the statistical significance of these changes could not be tested properly in this group. However, Luminex profiling, ELISA to determine IC₅₀ values and PCR for transcript levels all provided similar results; dexamethasone had the greatest effect on TNF-α and IL-6 production, while IL-8 was a consistently less Gc-sensitive cytokine.

Our findings with regard to IL-8 may be of clinical importance, as the levels of this neutrophil chemoattractant are increased in the airways of COPD patients [15], and are associated with decline in lung function [16]. Our data are compatible with Standiford et al. who reported that a high concentration of dexamethasone (1 µM) inhibited IL-8 production from healthy volunteer alveolar macrophages by only 61% [17], while in another study IL-8 production from healthy macrophages was not inhibited by 1 µM dexamethasone, in contrast to other cytokines [11]. Other studies have also reported that the effects of Gc on selected cytokines are modest, even at high concentrations in alveolar macrophages [13,18–20]. Additionally, Kent et al. profiled the effects of dexamethasone on inflammatory gene transcription in LPS-stimulated monocyte-derived macrophages from COPD patients and found a subset of genes that were Gc-insensitive, including IL-8 [13]. It is recognized that some macrophage inflammatory genes are Gc-insensitive [21] and a common theme of our current study and previous publications is that the alveolar macrophage production of IL-8, whether from healthy subjects or patients with disease, is less Gc-sensitive than other inflammatory mediators.

The possible molecular mechanisms responsible for the differences in Gc sensitivity between inflammatory genes are worthy of consideration. Many of the anti-inflammatory effects of Gc are due to transrepression, whereby the Gc/GR complex inhibits the function of transcription factors responsible for up-regulating inflammatory gene expression [22]. It is known that Gc do not target the whole inflammatory genome [21], as some genes are regulated by Gc-independent transcription mechanisms. IL-8 production is known to be dependent on transcriptional activation by NFκB and activator protein (AP)-1, and p38 mitogen-activated protein kinase (MAPK) stabilization of mRNA [23]. It is possible that the contribution of AP-1 and P38 MAPK signalling to IL-8 production in alveolar macrophages is Gc-insensitive.

We observed that cytokine production from unstimulated and LPS-stimulated alveolar macrophages was lower in COPD patients and S compared to HNS. This is similar to many previous publications showing that smoking reduces cytokine production from alveolar macrophages [24–30]. This was well demonstrated by Chen et al. in 25 smokers and 27 matched non-smokers, showing conclusively that the production of a range of cytokines from LPS-stimulated alveolar macrophages was reduced in smokers, associated with reduced MAPK and NFκB signalling [25]. This is compatible with observations that acute cigarette smoke exposure suppresses macrophage cytokine production [24,31]. In contrast, Cosio et al. reported that LPS-stimulated IL-8 and TNF-α production from alveolar macrophages was increased in COPD patients and S compared to HNS. The authors also reported that cytokine production was less sensitive to the effects of Gc in COPD patients and smokers [8]. The differences in the macrophage response to LPS in the Cosio study and the current report may be an underlying cause for altered pharmacological results observed when dexamethasone was used; Cosio et al. observed an increased LPS response with reduced Gc sensitivity in COPD, while we observed a decreased LPS response with no change in Gc sensitivity in COPD.

Culpitt et al. used cigarette smoke media to show that GM-CSF and IL-8 production from COPD alveolar macrophages is Gc-insensitive [7]. There is now growing evidence that cigarette smoke down-regulates many macrophage inflammatory genes, with the exception of IL-8 [24,25,31]. We therefore chose not to use cigarette smoke in the current study, as it would not have been possible to profile a range of cytokines and chemokines. It would have been of value to study other proinflammatory stimuli, including cytokines such as IL-1β or other Toll receptor ligands such as PolyIC. However, we were limited in the number of experiments that we could conduct due to the cell yield obtainable at bronchoscopy. Similarly, it would have been preferable to study other time-points in cell culture, to validate our data using a second Gc, and to study the full dose–response of dexamethasone in HNS, but cell yields were a limiting factor.

It is well known [5,6], and confirmed in this study, that COPD patients have increased numbers of macrophages in the airways. Ex-vivo cell cultures do not take these differences into account, as the same numbers of macrophages are placed into each culture well. We propose an alternative hypothesis for the involvement of alveolar macrophages in the Gc resistance observed in COPD patients in clinical practice; the production of some cytokines, including IL-8, from alveolar macrophages is relatively Gc-insensitive even in healthy subjects. In COPD patients, Gc are unable to suppress fully the production of these Gc-insensitive cytokines from the increased numbers of macrophages in the airways.

We did not observe any effect of current smoking status or ICS use in COPD patients on the LPS response or effect of dexamethasone in cell culture. This subanalysis was not statistically powered to evaluate such effects, but at least we can rule out any major effect of smoking status or ICS use on our results. The groups were not matched for gender, as we did not believe that this would not have influenced the results.

Our preliminary experiments optimized the cell culture conditions; we used a submaximal concentration of LPS, and chose to use 4 h as a time-point when protein production was near maximal and early gene expression changes could be studied. PCR analysis confirmed that there was substantial transcription induced by LPS at 4 h. Cell culture studies using monocytes/macrophages have used a variety of time-points, ranging from 4 to 24 h [7,8,11–13,16–30]. It would have been ideal to have studied a number of different time-points in the current study, for example both 4 and 24 h, but the limited number of cells available from bronchoscopy made this impossible.

Our BAL recovery data are compatible with previous findings of patients with emphysema having decreased BAL recovery volumes [32]. We did not quantify emphysema using computerized tomography scanning, but suggest that this is the reason for decreased cell counts in our COPD patients, some of whom probably had emphysema, compared to smokers.

In conclusion, we have identified some Gc-insensitive inflammatory mediators produced by LPS-stimulated alveolar macrophages, notably GM-CSF, G-CSF and IL-8. The Gc-insensitive production of these proteins is an intrinsic feature of healthy macrophages, and is also observed in those from COPD patients. We hypothesize that the increased numbers of macrophages in the lungs of COPD patients leads to an overall increase in the burden of cytokine production, including mediators such as IL-8, which are Gc-insensitive. These Gc-insensitive cytokines could be targeted with novel therapeutic approaches.

Disclosures

Jane Armstrong has no disclosures, Carol Sargent is an employee of AstraZeneca and Dave Singh has received lecture fees, research grants, consultancy fees and support conference attendance from various pharmaceutical companies including AstraZeneca, GlaxoSmithKline, Chiesi, Boehringer Ingleheim and Roche.