A Comparison of Serum and Sputum Inflammatory Mediator Profiles in Patients with Asthma and COPD

Abstract

Objective:

To investigate serum and sputum cytokine profiles in asthma and chronic obstructive pulmonary disorder (COPD).

Methods:

Clinical characteristics, blood and sputum samples were collected from asthma (n = 37) and COPD (n = 36) patients, and healthy subjects (n = 39). Cytokines were measured by enzyme-linked immunosorbent assay (ELISA) and inflammation antibody array; levels were analysed according to smoking status and presence of eosinophilia in the airways of COPD/asthma patients, and compared with those in healthy subjects.

Results:

ELISA revealed a significant difference in the expression of only a few cytokines in the COPD versus asthma groups, and in both patient groups compared with healthy subjects. The antibody array showed greater differentiation in inflammatory mediators. In the subphenotype analysis, the differential expression of cytokines was more significant between eosinophilic and noneosinophilic airway inflammation than between asthma and COPD, according to both the ELISA and antibody array. There was little difference in cytokine expression between smoking and nonsmoking subgroups.

Conclusions:

Differential expression of inflammatory mediators is present between patients with COPD or asthma, and eosinophilic or noneosinophilic airway inflammation, and contributes to understanding the cytokine network of airway inflammation.

Introduction

Traditionally, asthma and chronic obstructive pulmonary disorder (COPD) are considered to be diverse disease entities, with some clear physiological and anatomical extremes. Asthma, which is an allergic disease that develops in childhood, is characterized by variable and recurring symptoms, reversible airflow obstruction and bronchospasm. It has a generally good prognosis, with limited allergen exposure and anti-inflammatory treatment.^1,2 In contrast, COPD is most commonly caused by cigarette smoking, develops in middle or old age, and is physiologically characterized by airflow limitation that is incompletely reversible, so that lung function gets progressively worse and results in respiratory death.^{3 – 5} Cytokines play important roles in orchestrating airway inflammation in both asthma and COPD by recruiting, activating and regulating distinct types of inflammatory cells in the respiratory tract. ⁶ Patients with severe asthma have airway inflammation that is more comparable with that seen in COPD than in less-severe asthma. ⁷ In addition, over 40% of patients with COPD will be diagnosed with asthma, which demonstrates the overlapping features of both conditions.^{8 – 11} The correlation between asthma and COPD suggests a convergence of cytokine networks. Thus, studying a limited number of inflammatory mediators in asthma and COPD separately is of little significance.^7,12 It is important to consider them as a whole and recognize the similarity of cytokines in both diseases. Protein microarray technology offers a high-throughput platform for efficient profiling of gene expression and makes possible the simultaneous detection of the expression of multiple inflammatory mediators.^{13 – 15}

The present study compared serum and sputum cytokine profiles in outpatients with asthma and COPD, and in healthy subjects, using enzyme-linked immunosorbent assay (ELISA) and protein microarray technology. In addition, inflammatory mediator expression was investigated in subphenotypes based on smoking status and presence of eosinophilic airway inflammation.

Patients and methods

Study Population

Consecutive adults with asthma and COPD were recruited from outpatients attending Huashan Hospital, Fudan University, Shanghai, China, between August 2009 and March 2011. Healthy subjects, recruited from the hospital (such as members of staff) and other sources (such as local schools), were included as controls. A diagnosis of asthma was made by a respiratory physician, based on the guidelines of the Global Initiative for Asthma (GINA; http://www.ginasthma.org/), where there was consistent symptomatology. All COPD patients who had airflow obstruction with a postbronchodilator forced expiratory volume in 1s (FEV1)/forced vital capacity (FVC) ratio (FEV1/FVC) < 0.7, and who demonstrated bronchodilator reversibility, were included in the study. Patients with a diagnosis of acute exacerbation of COPD and asthma, active pulmonary tuberculosis, other clinically relevant lung or medical illnesses or the inability to produce sputum following the induced sputum procedure were excluded from the study. Age- and sex-matched healthy control subjects were defined on the basis of lack of clinical asthmatic or COPD history (or other clinically relevant lung or medical illnesses), no upper respiratory tract infection within the previous 2 weeks, and no medication usage.

The clinical trial protocol was approved by the Ethics Committee of Huashan Hospital, and written informed consent was obtained from each participant.

Study Design

Demographic data including age, duration of symptoms and full smoking history was recorded for each participant. Spirometry was performed on all participants according to the American Thoracic Society and European Respiratory Society guidelines, ¹⁶ and FEV1 and FVC were measured. Blood and sputum were collected for measuring levels of inflammatory cells and mediators. Asthma and COPD patients were classified by eosinophilic or noneosinophilic airway inflammation according to baseline sputum eosinophil counts, and by smoking status; the groups compared well with each other and with healthy subjects. Sputum eosinophilia was defined as eosinophils > 2% of nonsquamous cells. ¹⁷

Sputum Induction and Processing

The process of sputum induction was conducted as described by Pin et al. ¹⁸ After the measurements of baseline FEV1, patients were pretreated with inhaled salbutamol (two sprays, ∼200 μl; 8.36 mmol/l) using a metered-dose inhaler. After 10 min, patients inhaled a nebulized solution of hypertonic sterile saline (5%) over three 5-min periods using an ultrasonic nebulizer (Omron compression atomizer, NE-C29; Omron Healthcare UK, Milton Keynes, UK) at a spray rate of 0.4 ml/min, particle size (mass diameter) < 5 μm. Patients were asked to rinse their mouth with water and were then instructed to cough sputum into a sterilized sputum container. This procedure continued until the FEV1 had fallen to ≥ 20% or an adequate sputum sample (∼1 ml) was obtained.

Sputum was extracted from saliva and processed as described previously. ¹⁹ Sputum was treated with 0.1% dithiothreitol (volume equal to four times the weight of collected sputum) and mixed by rotating for 30 min at 37°C, followed by the addition of a fourfold volume of 0.01 M phosphate-buffered saline (PBS; pH 7.4) and rotation for a further 30 min. The suspension was filtered using a 0.1-mm filter (G&G Precise Electronic Co. Ltd, Shanghai, China), then centrifuged at 2054 g for 20 min. The supernatant (∼2 ml) was aspirated and stored at –80°C until analysis of inflammatory mediators. Cytospins were prepared and stained with May –Grünwald–Giemsa (1: 5 stock solution), a differential cell count of 400 nonsquamous cells was performed and eosinophils were expressed as a percentage of the total nonsquamous cells.

Serum Preparation

Venous blood (5 ml) was collected using commercial collection tubes (BD Biosciences, San Jose, CA, USA) and allowed to stand for 1 - 2 h at room temperature before being centrifuged for 10 min at 2683 g. Serum (1 ml) was aspirated, aliquoted into Eppendorf tubes and stored at -80°C until analysis.

Assessment of Inflammatory Mediators by Elisa

Levels of interleukin (IL)-4, IL-5, IL-6, IL-8 and IL-10, leukotriene B4 (LTB-4), C-reactive protein, tumour necrosis factor (TNF)-α, interferon (IFN)-γ, and transforming growth factor (TGF)-β were evaluated in serum samples. IL-4, IL-5, IL-6, IL-8, IL-10, LTB-4, TNF-α, IFN-γ and TGF-β were also evaluated in sputum supernatant, by ELISA (R&D Systems, Minneapolis, MN, USA), according to the manufacturer's instructions. Data were expressed as pg/ml.

Assessment of Inflammatory Mediators by Antibody Array

A biotin label-based cytokine antibody array was used to examine simultaneously the levels of 40 inflammatory mediators in sera from healthy volunteers, and from asthma and COPD patients. This included the following inflammatory mediators: IL-1α; IL-1β; IL-2; IL-3; IL-4; IL-6; IL-6 soluble receptor (IL-6sR); IL-8; IL-10; IL-15; TNF-α; TNF-β; soluble TNF receptor (sTNFR) I; sTNFRII; IFN-γ; monocyte chemoattractant I-309; intercellular adhesion molecule (ICAM)-1; monocyte chemotactic protein (MCP)-1; MCP-2; macrophage inflammatory protein (MIP)-1α; MIP-1β; MIP-1δ; regulated on activation, normal T cell expressed and secreted (RANTES); eosinophil chemotactic protein (EOTAXIN); EOTAXIN-2; tissue inhibitor of metalloproteinases (TIMP)-2.

Serum protein levels were analysed with the RayBio™ Human Inflammation Antibody Array III (Ray Biotech, Norcross, GA, USA) according to the manufacturer's instructions. Briefly, the arrays (membranes) were blocked with 5% bovine serum albumin/Tris-buffered saline (TBS; 0.01 M Tris-HCl, pH 7.6, 0.15 M NaCl) for 1 h at room temperature. The blocked membranes were then incubated with ∼2 ml of conditioned media, provided with the RayBio™ Human Inflammation Antibody Array III kit, or 1 ml of the participant's serum, after normalization for equal amounts of protein. After extensive washing with 3 ml TBS/0.1% v/v Tween 20 for three 5-min periods and 3 ml TBS for two 5-min periods to remove unbound materials, the membranes were incubated for 1 h at room temperature with a cocktail of biotin-labelled antibodies against the different individual inflammatory mediators mentioned above. The membranes were then washed and incubated with 4 ml horseradish peroxidase (HRP)-conjugated streptavidin (2.5 pg/ml) for 1 h at room temperature. Unbound HRP-streptavidin was washed three times with 3 ml TBS/0.1% v/v Tween 20 for three 5-min periods. Finally, the signals were detected using enhanced chemiluminescence reagents provided with the RayBio™ Human Inflammation Antibody Array III kit. The net density grey level for each inflammatory mediator was determined by subtracting the background from the total raw density grey level. The relative fold difference represented the amount of cytokine/inflammatory mediator present on the control culture membrane, which was based on the ratio of the mean grey levels in the patient samples to the mean grey levels in control subject samples. All the protein antibody arrays were performed by the Shanghai Kangcheng Biotech Company, Shanghai, China.

Statistical Analyses

Data were analysed using the SPSS^® software package, version 16.0 (SPSS Inc., Chicago, IL, USA) for Windows^®. Results were reported as mean ± SE. Analysis was performed using Student's t-test and analysis of variance for parametric data, the two-sample Wilcoxon rank–sum test or Kruskal–Wallis test for nonparametric data, and Pearson's χ²-test for categorical data. A P-value < 0.05 was considered to be statistically significant.

Results

The present study included 112 participants: patients with asthma (n = 37) or COPD (n = 36) and healthy volunteers (n = 39). Baseline clinical characteristics of all participants are presented in Table 1. A total of 103 patients tolerated the sputum induction procedure; four patients with severe persistent asthma (according to the GINA criteria) developed bronchospasm, which was quickly reversed by salbutamol (two sprays, ∼200 μl; 8.36 mmol/l) nebulization. Five subjects were unable to produce a suitable quantity of sputum. There was a significant difference between patients with COPD compared with those with asthma and healthy subjects with respect to age; there were no significant between-group differences in terms of sex, disease duration and dosage of inhaled corticosteroids.

Table 1:

Baseline clinical characteristics of patients with asthma or chronic obstructive pulmonary disease (COPD) and healthy subjects

Parameter	Asthma patients n = 37	COPD patients n = 36	Healthy subjects n = 39	Statistical significance b
Age, y	47.48 ± 2.31	66.86 ± 1.71 a	48.53 ± 1.73	P < 0.01
Males	18 (48.6)	18 (50.0)	20 (51.3)	NS
Smokers	4 (10.8) a	16 (44.4) a	0 (0)	P < 0.01
ICS	18 (48.6)	21 (58.3)	-	NS
Disease duration, y	9.41 ± 2.20	15.50 ± 2.36	-	NS
FEV1, % predicted	75.55 ± 4.83 a	42.76 ± 4.46 a	102.86 ± 1.91	P < 0.01
FEV1/FVC, %	98.66 ± 2.39	72.76 ± 4.46 a	103.14 ± 2.58	P < 0.01
EA	18 (48.6)	10 (27.8)	-	NS

Data presented as n (%) of patients or mean ± SE.

a

P < 0.05 versus healthy subjects;

b

comparison among the three study groups; Student's t-test and analysis of variance for parametric data, two-sample Wilcoxon rank–sum test or Kruskal–Wallis test for nonparametric data and Pearson's χ²-test for categorical data.

NS, not statistically significant (P ≥ 0.05).

ICS, inhaled corticosteroids; FEV1, forced expiratory volume in 1 s; FEV1% predicted, FEV1/FVC ratio of the patient divided by the average FEV1% in the population for any person of similar age, sex and body composition; FVC, forced vital capacity; EA, eosinophilia (> 2% of nonsquamous cells).

As determined by ELISA, serum inflammatory mediator profiles were comparable between asthma and COPD patients with the exception of IL-8, which was significantly increased in the COPD group versus the asthma group (P < 0.05). The level of IL-8 in serum, and levels of TGF-β, IL-8 and LTB-4 in sputum, were also significantly elevated in COPD patients compared with healthy subjects (P < 0.05; Fig. 1A). There was no difference in inflammatory mediator expression between asthma patients and healthy subjects.

Figure 1:

Expression of inflammatory mediators in serum (sr) and sputum (sp) samples from asthma (n = 37) or chronic obstructive pulmonary disease (COPD) (n = 36) patients, and from healthy subjects (n = 39). Inflammatory mediators measured by (A) enzyme-linked immunosorbent assay and (B) inflammation antibody array. ^aAsthma versus COPD patients, ^bCOPD patients versus healthy subjects (P < 0.05); analyses performed by the Student's t-test and analysis of variance for parametric data, or the two-sample Wilcoxon rank–sum test or Kruskal–Wallis test for nonparametric data. Data presented as mean ± SE. Inflammatory mediators are defined in the text

According to antibody array analyses, with the exception of IL-4 levels (which were significantly higher in asthma patients compared with COPD patients), IL-1α, IL-8, IL-15, MCP-1, MIP-1δ and sTNFRII levels were all significantly lower in asthma patients than in COPD patients (P < 0.05; Fig. 1B). Compared with healthy subjects, levels of IL-3 and IL-10 in asthma patients and I-309, IL-10, IL-15 and TNF-β in COPD patients were significantly lower, while levels of RANTES and EOTAXIN-2 in asthmatic patients, and IL-1α, IL-6sR, IL-8, MIP-1β, MIPδ, RANTES and sTNFRII in COPD patients, were significantly higher (P < 0.05; Fig. 1B). There was no significant difference in the expression of 14 other inflammatory mediators among asthma or COPD patients and healthy subjects.

Differential expression of cytokines and chemokines was observed among COPD and asthma patients with eosinophilic (n = 28) or noneosinophilic (n = 45) airway inflammation and healthy subjects (n = 39). As shown by ELISA, levels of TNF-α, TNF-β, IL-6 and IL-8 were increased, whereas IL-5 levels were decreased, in sputum samples from patients with noneosinophilic airway inflammation compared with patients with eosinophilic airway inflammation (P < 0.05; Fig. 2A). Elevated TNF-β and IL-5 levels were also observed in sputum samples from patients with eosinophilic and noneosinophilic airway inflammation respectively, while IL-8 levels declined in both groups compared with the respective cytokine levels in healthy subjects (P < 0.05; Fig. 2A). As shown by inflammation antibody array, levels of IL-6, IL-8, sTNFRI and sTNFRII in patients with noneosinophilic airway inflammation were higher than those in patients with eosinophilic airway inflammation, and there was no lower expression of any inflammatory mediator in the patients with noneosinophilic airway inflammation. Compared with healthy subjects, levels of I-309, IL-3, IL-10, IL-15 and TNF-β in patients with noneosinophilic airway inflammation, and IL-3, IL-6, IL-15 and TNF-β in patients with eosinophilic airway inflammation, were lower; levels of IL-8, sTNFRI, sTNFRII and TIMP-2 in patients with noneosinophilic airway inflammation, and RANTES in patients with eosinophilic and noneosinophilic airway inflammation, were higher (P < 0.05; Fig. 2B).

Figure 2:

Comparison of inflammatory mediators in serum (sr) and sputum (sp) in patients with chronic obstructive pulmonary disease (COPD) or asthma, stratified by eosinophilic (n = 28) or noneosinophilic airway inflammation (n = 45), and in healthy subjects (n = 39). Inflammatory mediators measured by (A) enzyme-linked immunosorbent assay and (B) inflammation antibody array. ^aNoneosinophilic airway inflammation versus eosinophilic airway inflammation; ^bpatients with eosinophilic or noneosinophilic airway inflammation versus healthy subjects (P < 0.05); analyses were performed by Student's t-test and analysis of variance for parametric data, or the two-sample Wilcoxon rank–sum test or Kruskal–Wallis test for nonparametric data. Data presented as mean ± SE. Inflammatory mediators are defined in the text

There was differential expression of inflammatory mediators in COPD or asthma patients stratified as smokers (n = 20) or nonsmokers (n = 53), and healthy subjects (n = 39) (Fig. 3). Using the ELISA assay, compared with healthy subjects, LTB-4 levels in the sputum of smokers were significantly lower, and IL-5 levels in the sputum of smokers and nonsmokers were significantly higher (P < 0.05; Fig. 3A). Compared with healthy subjects, as shown by inflammation antibody array, serum levels of IL-3, IL-10 and IL-15 in smokers and nonsmokers were significantly lower, whereas the level of RANTES was significantly higher. In addition, the TNF-β level in nonsmokers and smokers was significantly lower than that in healthy subjects (P < 0.05; Fig. 3B). There was no significant difference in the expression of 14 other inflammatory mediators, among smokers and nonsmokers with asthma or COPD and healthy subjects.

Figure 3:

Comparison of inflammatory mediators in serum (sr) and sputum (sp) samples from chronic obstructive pulmonary disease (COPD) and asthma patients stratified by smokers (n = 20) and nonsmokers (n = 53) with airway inflammation, and in healthy subjects (n = 39). Inflammatory mediators measured by (A) enzyme-linked immunosorbent assay and (B) inflammation antibody array. ^aNonsmokers versus smokers, ^bsmokers or nonsmokers versus healthy subjects (P < 0.05); analyses were performed by the Student's t-test and analysis of variance for parametric data, or the two-sample Wilcoxon rank–sum test or Kruskal–Wallis test for nonparametric data. Data presented as mean ± SE. Inflammatory mediators are defined in the text

Discussion

Asthma and COPD have long been considered to be separate disease entities due to their different clinical phenotypes. There are, however, similarities in the types of inflammatory cells observed in the airways of patients with these diseases, ²⁰ and cytokines and chemokines secreted by these types of cell interact as a network of inflammatory mediators. Considering the connection between cytokines and chemokines underlying COPD and asthma, it is necessary to define the subphenotypes of inflammatory mediators.

The present study evaluated a panel of cytokines and chemokines, and showed that, when assessed by ELISA, there was little variation between asthma and COPD patients (albeit that levels of a few inflammatory mediators were higher in COPD patients compared with healthy subjects). Assessment by inflammation antibody array showed a significant difference in the expression profile of several inflammatory mediators between patients with asthma and COPD, and between patients with or without eosinophilic airway inflammation. Some of these alterations are well established, such as elevated IL-8 in the serum and sputum of COPD subjects^21,22 and increased IL-4 in the serum of asthma subjects.^23,24 Other proven cytokine alterations between asthma and COPD, such as changes in IL-5, were not shown in the present study, although research has previously demonstrated that IL-5 levels are much higher in asthma patients compared with COPD patients. ²⁵ In addition, the significance of other changes in inflammatory mediators among asthma and COPD patients and healthy subjects needs further investigation. For example, levels of IL-1α in people with COPD were significantly higher than in asthmatics and healthy controls. IL-1α has been identified as a mediator of smoke-induced neutrophilic inflammation in a murine model of COPD, ²⁶ whereas its role in asthma is seldom referred to in the literature.

Advances in knowledge related to the pathogenesis and clinical characteristics of airway inflammatory diseases (such as asthma and COPD) has led to an awareness of the ambiguity between these two diseases, with overlapping phenotypes that make it difficult to distinguish between both conditions. Thus, the development of a new taxonomy is required.^27,28 The present study considered asthma and COPD both as separate entities and together, but stratified by the presence or absence of eosinophilic airway inflammation or by smoking status. When airway inflammation was reclassified on the basis of eosinophilic and noneosinophilic subphenotypes, the differences between inflammatory mediators were more pronounced than the differences between asthma and COPD, both in the results measured by ELISA and by inflammation antibody array. Eosinophilic and noneosinophilic phenotypes were not, however, closely correlated with the diagnosis of asthma and COPD, since the proportion of patients with eosinophilic airway inflammation was not significantly different between the two patient groups. This may in turn indicate the importance of stratifying for inflammatory airway subphenotypes. Tobacco smoking is believed to be one of the environmental factors that affects the prevalence of COPD and bronchial asthma. ²⁹ When COPD and asthma patients were reclassified according to smoking status, the differential expression of inflammatory mediators between smokers or nonsmokers with airway inflammation was similar to the difference between asthma and COPD. This is probably because only 10.8% of asthmatic patients in the present study were smokers. Studies including more participants should be carried out, in order to determine the role of smoking on the taxonomy of airway inflammation.

The present study employed an inflammation antibody array, which is a highly sensitive approach for detecting levels of multiple cytokine and chemokine expression from patients' serum, simultaneously.^{30 – 32} Traditionally, cytokines and chemokines are detected by ELISA; however, the antibody array approach has advantages over ELISA. The differences in inflammatory mediators detected by the antibody array were much greater than those measured by ELISA. Furthermore, the sensitivity and detection range of the protein antibody array was considerably increased. Thus, the present study supports the use of simple, flexible, high-throughput (but inexpensive) inflammation antibody array technology to compare cytokine and chemokine profiles in serum samples from patients with asthma or COPD with those from healthy subjects, and represents a valuable investigational tool to identify potential differential expression involved in airway inflammation.

The cytokine profiles in the cultures of blood mononuclear cells from patients with COPD and bronchial asthma have been reported previously. ³³ Both sputum and serum samples were collected in the present study and, in general, the differences between inflammatory mediators in the sputum from COPD or asthma patients were more noteworthy than those observed in serum. This may indicate that sputum samples are more sensitive than serum for examination of the airway inflammatory response. It was, however, not possible to investigate cytokine and chemokine expression in sputum using the inflammation antibody array, due to the limited quantity of patients' sputum collected. Another limitation was the relatively small sample size; inclusion of a greater number of patients would provide a more accurate array picture.

In conclusion, the present study indicates that both traditional descriptive disease terms (asthma and COPD) and new taxonomy (eosinophilic and noneosinophilic airway inflammation), show differential expression of inflammatory mediators, which is particularly obvious in sputum samples. Evaluation of these two types of categorization appears to be complementary. It also increases our understanding of the cytokine and chemokine network of airway inflammation and, potentially, its subsequent diagnosis or management.